Next Issue
Volume 135, CSDU-CSSI DAYS 25
Previous Issue
Volume 133, EASN 2025
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
1.2
Journals
Engineering Proceedings
Volume 134
Issue 1
engproc-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Eng. Proc., 2026, ECICE 2025

The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025)

Yunlin, Taiwan | 14–16 November 2025

Volume Editors:
Teen-Hang Meen, Department of Electronic Engineering, National Formosa University, Yunlin, Taiwan
Chi-Ting Ho, Department of Mechanical Design Engineering, National Formosa University, Huwei, Yunlin, Taiwan
Cheng-Fu Yang, Department of Chemical and Materials Engineering, National University of Kaohsiung, Kaohsiung, Taiwan

Number of Papers: 93
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Cover Story (view full-size image): The 2025 IEEE 7th Eurasia Conference on IoT, Communication and Engineering (ECICE 2025) was held in Yunlin, Taiwan, from 14 to 16 November 2025, and it provided a unified communication platform for [...] Read more.
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Other

6 pages, 530 KB  
Proceeding Paper
Classification of Guava Leaf Disease Using Support Vector Machine and You Only Look Once Version 8
by Paul Jess C. Rosero, Frances Mae P. Domingo and Analyn N. Yumang
Eng. Proc. 2026, 134(1), 1; https://doi.org/10.3390/engproc2026134001 - 26 Mar 2026
Viewed by 4100
Abstract
Guava is a popular fruit in the Philippines, as it offers various health benefits. Its leaves are used in traditional medicine to aid in wound healing, stomach disorders, pain relief, and more. In this study, we classified guava leaf diseases using Support Vector [...] Read more.
Guava is a popular fruit in the Philippines, as it offers various health benefits. Its leaves are used in traditional medicine to aid in wound healing, stomach disorders, pain relief, and more. In this study, we classified guava leaf diseases using Support Vector Machine (SVM) and You Only Look Once version 8 (YOLOv8). Raspberry Pi 4 is used to control the image preprocessing and the program that utilizes the proposed trained model. The SVM model conducts image classification, while YOLOv8 handles feature extraction and object detection. Grayscale and color thresholding segmentation feature extraction is also implemented in the proposed model. The developed model combines both YOLOv8 and SVM algorithms to evaluate their accuracy using a confusion matrix, achieving a 92.5% accuracy. With its very low error rate, the system can accurately classify guava leaf diseases. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 873 KB  
Proceeding Paper
Utilizing Residual Network 50 Convolutional Neural Network Architecture for Enhanced Philippine Regional Language Classification on Jetson Orin Nano
by John Paul T. Cruz, Aaron B. Abadiano, FP O. Sangilan, Emmy Grace T. Requillo and Roben A. Juanatas
Eng. Proc. 2026, 134(1), 2; https://doi.org/10.3390/engproc2026134002 - 26 Mar 2026
Viewed by 408
Abstract
Visual speech recognition systems encounter significant challenges in multilingual nations such as the Philippines, where numerous regional languages, including Cebuano and Ilocano, feature distinct phonetic-visual characteristics. Deep learning models such as the Lip Reading Network and the Lightweight Crowd Segmentation Network have demonstrated [...] Read more.
Visual speech recognition systems encounter significant challenges in multilingual nations such as the Philippines, where numerous regional languages, including Cebuano and Ilocano, feature distinct phonetic-visual characteristics. Deep learning models such as the Lip Reading Network and the Lightweight Crowd Segmentation Network have demonstrated strong performance with 3D Convolutional Neural Networks (CNNs). However, their substantial computational requirements restrict deployment on portable edge devices. We introduce a more efficient alternative that integrates a 2D Residual Network 50 architecture with a Long Short-Term Memory network and Connectionist Temporal Classification for lip-reading classification of Philippine regional languages. The proposed model is deployed on the Jetson Orin Nano, a high-performance edge device optimized for real-time inference through Compute Unified Device Architecture acceleration. Using a dataset of 2000 annotated videos encompassing 10 lexicons each for Cebuano and Ilocano, the model’s effectiveness was evaluated. Results achieved a regional language classification accuracy of 90%, with lexicon-level accuracies of 74% for Cebuano and 66% for Ilocano. This work represents a step toward developing accessible and scalable communication aids for deaf communities in linguistically diverse environments, leveraging transfer learning on pretrained models. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 779 KB  
Proceeding Paper
Evaluation of Convolutional Neural Network Models for Leaf-Based Endangered Tree Classification
by Bem Gumapac and Jocelyn Villaverde
Eng. Proc. 2026, 134(1), 3; https://doi.org/10.3390/engproc2026134003 - 27 Mar 2026
Viewed by 517
Abstract
The conservation of endemic tree species in the Philippines requires lightweight and localized technological solutions. Therefore, we benchmark three Convolutional Neural Network (CNN) architectures: MobileNetV2, EfficientNetB0, and NASNetMobile, using 1800 leaf images from five endangered Dipterocarpaceae species and an unknown class. The models [...] Read more.
The conservation of endemic tree species in the Philippines requires lightweight and localized technological solutions. Therefore, we benchmark three Convolutional Neural Network (CNN) architectures: MobileNetV2, EfficientNetB0, and NASNetMobile, using 1800 leaf images from five endangered Dipterocarpaceae species and an unknown class. The models were trained with transfer learning and deployed on a Raspberry Pi prototype. Validation accuracies exceeded 97.00%, with NASNetMobile reaching 99.44%. Field testing produced lower accuracies of 85.00% for NASNetMobile, 77.78% for MobileNetV2, and 66.11% for EfficientNetB0. One-way analysis of variance with Tukey’s Honestly Significant Difference showed no statistically significant differences at the 0.05 level. These findings highlight the importance of field validation and statistical testing in CNN benchmarking for embedded biodiversity monitoring. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

12 pages, 1606 KB  
Proceeding Paper
Finite Impulse Response Digital Filter Implementation Using Quantum Computation and Orthogonal Triangular Decomposition
by Chien-Cheng Tseng and Su-Ling Lee
Eng. Proc. 2026, 134(1), 4; https://doi.org/10.3390/engproc2026134004 - 27 Mar 2026
Viewed by 348
Abstract
In digital signal processing, the finite impulse response (FIR) filter is a fundamental tool for processing discrete-time signals. This paper explores the implementation of FIR filters using quantum computation methods. In this study, a quantum circuit for the FIR filter is designed using [...] Read more.
In digital signal processing, the finite impulse response (FIR) filter is a fundamental tool for processing discrete-time signals. This paper explores the implementation of FIR filters using quantum computation methods. In this study, a quantum circuit for the FIR filter is designed using a normalized filter coefficient vector, orthogonal triangular decomposition commonly known as QR decomposition, and the transpilation tools provided by IBM’s software Qiskit SDK V2.3. Then, each block of the input signal is normalized to a unit-norm vector, loaded into a quantum register, and processed by the FIR filter quantum circuit to produce an output state. Quantum measurement is then performed on the output state to obtain a histogram, from which the first-bin data are scaled to compute the output sample of the filter. Finally, signal filtering experiments using FIR mean filters are conducted to demonstrate the effectiveness of the proposed quantum computation approach. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

5 pages, 622 KB  
Proceeding Paper
Improvement of an Automated Process for Folding Soft Plastic Bag
by Kai-Yuan Huang and Kuang-Chyi Lee
Eng. Proc. 2026, 134(1), 5; https://doi.org/10.3390/engproc2026134005 - 27 Mar 2026
Viewed by 274
Abstract
We developed an automated folding device for soft plastic bags to replace manual folding. The device employs a flat folding plane, combined with feeding positioning and edge-guided pre-crease. After computing the required torque for every folding step, the actuators are selected. This device [...] Read more.
We developed an automated folding device for soft plastic bags to replace manual folding. The device employs a flat folding plane, combined with feeding positioning and edge-guided pre-crease. After computing the required torque for every folding step, the actuators are selected. This device increases production output and reduces labor costs. Under typical operating conditions, the production of folded bags has increased to 90 bags per hour, three times that of manual folding, while controlling the crease position deviation within 2.0 mm. With one automated folding device, instead of one laborer, the yield rate was raised to 98%, the production was 187,200 bags per year, and the annual savings were estimated as NT$240,000. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 866 KB  
Proceeding Paper
Inspection for Solder Joint Defects in Voltage Regulator ICs of Automotive Charging Applications
by Yi-Hsuan Chiu and Kuang-Chyi Lee
Eng. Proc. 2026, 134(1), 6; https://doi.org/10.3390/engproc2026134006 - 27 Mar 2026
Viewed by 317
Abstract
In automated production lines for automotive chargers, solder joint inspection is critical due to the widespread adoption of automotive electronics and electric vehicles. This study establishes a You Only Look Once Version 8 (YOLOv8)-based single-pin solder joint classification model for an 8-pin automotive [...] Read more.
In automated production lines for automotive chargers, solder joint inspection is critical due to the widespread adoption of automotive electronics and electric vehicles. This study establishes a You Only Look Once Version 8 (YOLOv8)-based single-pin solder joint classification model for an 8-pin automotive voltage regulator IC. Solder joints were categorized into four types: normal, misalignment, insufficient fillet, and cold joint. The model achieved a single-pin training accuracy of 0.987 (4000 samples) and a test accuracy of 0.973 (4800 samples), while overall IC-level evaluation exceeded 0.90. Normal and cold joint categories were detected with the highest reliability, whereas occasional misclassifications occurred in the insufficient fillet and misalignment categories. These results demonstrate that the proposed method is feasible for efficient and accurate detection of solder joint defects, providing a practical approach to support automated inspection and ensure consistent production quality. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

13 pages, 1799 KB  
Proceeding Paper
Cooling Tower Decision Support Web System: A Case Study
by Hao-Yu Lien, Wen-Hao Chen and Yen-Jen Chen
Eng. Proc. 2026, 134(1), 7; https://doi.org/10.3390/engproc2026134007 - 30 Mar 2026
Viewed by 420
Abstract
Conventional cooling tower operations often rely on the operator’s experience for fan-switching control, lacking precise decision support and real-time monitoring capabilities. This makes it challenging to maintain water temperature within an optimal range, thereby affecting industrial process efficiency. Using a case study approach, [...] Read more.
Conventional cooling tower operations often rely on the operator’s experience for fan-switching control, lacking precise decision support and real-time monitoring capabilities. This makes it challenging to maintain water temperature within an optimal range, thereby affecting industrial process efficiency. Using a case study approach, we integrate a Long Short-Term Memory (LSTM) model for temperature prediction with a Reinforcement Learning (RL) model to develop a web-based decision support system for cooling tower operations. The system uses an LSTM model to predict the trend of return water temperature for the next 15 min. This prediction, along with environmental conditions and historical data, is then fed into the RL model. Through a reward mechanism, the model is designed to receive a higher score when the predicted temperature is close to the benchmark of 30.5 °C and a lower score otherwise, enabling it to learn the optimal fan control strategy. Based on the evaluation results, the system automatically determines the optimal action—turning the fan on, off, or maintaining its current state—and provides specific fan operation suggestions and a decision-making basis to the operator via a web interface. This system is designed with a layered architecture, comprising functional modules such as a real-time monitoring dashboard, historical data query, and AI model management. Through visual elements like temperature trend line charts, fan status indicators, and a decision suggestion interface, it provides operators with real-time water temperature status, predicted temperature trends, and specific operational recommendations. The system has been deployed and is running in an actual manufacturing factory, where the AI model generates predictions and decision outputs every 15 min, assisting operators in adjusting fan control. This has successfully stabilized the outlet water temperature within the target range of 30–31 °C, thereby enhancing the efficiency of cooling water temperature regulation. The model presents the practical application of AI technology in a manufacturing control scenario and establishes a web-based decision support system, providing a concrete example for smart manufacturing transformation within an Industrial IoT environment. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 1451 KB  
Proceeding Paper
Time-Sensitive Networking and Time Scheduling Mechanisms for 5G Networks
by Po-Kai Chuang, Ming-Hung Lee, Yu-Chuan Luo, Jian-Kai Huang, Chin-Cheng Hu and Yu-Ping Yu
Eng. Proc. 2026, 134(1), 8; https://doi.org/10.3390/engproc2026134008 - 30 Mar 2026
Viewed by 387
Abstract
With the rapid development of 5G communication technology, 5G networks are designed to achieve three major objectives: higher bandwidth, support for a greater number of connected devices, and lower latency. It is necessary to meet the requirements of the three primary 5G application [...] Read more.
With the rapid development of 5G communication technology, 5G networks are designed to achieve three major objectives: higher bandwidth, support for a greater number of connected devices, and lower latency. It is necessary to meet the requirements of the three primary 5G application scenarios: Enhanced Mobile Broadband, Massive Machine-Type Communications, and Ultra-Reliable and Low Latency Communications (uRLLC). To meet the stringent requirements for time synchronization and low latency, 5G is being integrated with Ethernet-based Time-Sensitive Networking (TSN) technologies. TSN plays an important role in achieving time determinism in uRLLC scenarios and ensures low-latency and high-reliability Ethernet communication through the transmission of time signals that are also known as the Precision Time Protocol. We applied TSN technology in the Institute of Electrical and Electronics Engineers 802.1Qbv standard and evaluated its transmission delay performance. Modifying the gate control list (GCL) to accommodate varying network traffic ensures low-latency transmission for high-priority traffic. We propose two GCL configurations for TSN that incorporate time-aware shaper to achieve efficient traffic scheduling. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 2729 KB  
Proceeding Paper
Unmanned Aerial Vehicles Aerial Photography Combined with Building Information Modeling Applied in Road Landscape Planning Research
by Ren-Jwo Tsay
Eng. Proc. 2026, 134(1), 9; https://doi.org/10.3390/engproc2026134009 - 30 Mar 2026
Viewed by 316
Abstract
In road planning and landscape design, data collection emphasizes existing site conditions, particularly in projects involving modifications rather than new construction, as such data directly inform subsequent planning decisions. Beyond conventional surveying techniques, large-scale street-region digital elevation models can be generated using aerial [...] Read more.
In road planning and landscape design, data collection emphasizes existing site conditions, particularly in projects involving modifications rather than new construction, as such data directly inform subsequent planning decisions. Beyond conventional surveying techniques, large-scale street-region digital elevation models can be generated using aerial imagery acquired from unmanned aerial vehicles. The point clouds derived from these aerial photographs provide a basis for constructing spatial models applicable to street landscape and road planning. In this study, aerial data were processed using Pix4D software 4.8.4 to generate the initial spatial model, which was subsequently integrated into a building information modeling-based design framework in Autodesk Revit 2022. This approach enabled rapid and precise design outputs, while the resulting BIM model was further applied to mapping applications to establish a foundational database for regional public works. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 2523 KB  
Proceeding Paper
AI- and IoT-Enabled Smart Dustbin for Automated Hazardous Electronic Waste Separation
by Min Xuan Soh, Hou Kit Mun, Hui Ziang Lee, Zhi Khai Ng and Yan Chai Hum
Eng. Proc. 2026, 134(1), 10; https://doi.org/10.3390/engproc2026134010 - 30 Mar 2026
Viewed by 590
Abstract
Electronic waste (e-waste) continues to increase globally, yet conventional bins cannot distinguish hazardous batteries and devices from recyclable metals. This article presents an AI- and IoT-enabled smart dustbin that automatically identifies and segregates general waste, metals, and electronic or battery-based hazards while providing [...] Read more.
Electronic waste (e-waste) continues to increase globally, yet conventional bins cannot distinguish hazardous batteries and devices from recyclable metals. This article presents an AI- and IoT-enabled smart dustbin that automatically identifies and segregates general waste, metals, and electronic or battery-based hazards while providing real-time monitoring through a cloud-based dashboard. The system integrates inductive sensing, Time-of-Flight detection, an Espressif Systems Platform 32 (ESP32)-CAM module, and Google Gemini 1.5 Flash for image classification. The prototype achieved a waste segregation accuracy of 93.5% with a total cycle time of 4–6 s per item. The touch-free lid, swift mechanical actuation, and compact 59 × 59 × 100 cm footprint make the dustbin suitable for deployment in campuses, offices, and shopping malls. Dual ESP32 controllers, cloud connectivity through Message Queuing Telemetry Transport (MQTT), Firebase, and a Streamlit web interface enable automated alerts through Discord and email, demonstrating a scalable and energy-efficient approach to sustainable e-waste management. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1880 KB  
Proceeding Paper
Closed-Loop Personal Protective Equipment Compliance System
by Kuan-Chun Huang, Mathieu Bodin, Hsiao-Tse Lin, Wei-Nung Huang and Hsiang-Yu Wang
Eng. Proc. 2026, 134(1), 11; https://doi.org/10.3390/engproc2026134011 - 30 Mar 2026
Viewed by 309
Abstract
We developed a Python-integrated closed-loop industrial safety system that bridges real-time helmet-compliance detection with immediate machine control. The custom Python application serves as critical middleware, orchestrating the complete pipeline from You Only Look Once Version 8 computer vision inference to industrial automation by [...] Read more.
We developed a Python-integrated closed-loop industrial safety system that bridges real-time helmet-compliance detection with immediate machine control. The custom Python application serves as critical middleware, orchestrating the complete pipeline from You Only Look Once Version 8 computer vision inference to industrial automation by trans-lating AI detection results into Object Linking and Embedding for Process Control Unified Architecture communications with a Mitsubishi programmable logic controller (PLC). The Python framework implements configurable safety policies through polygonal zones with authorized helmet colors, incorporates persistence filtering to prevent nuisance trips, and ensures deterministic translation from probabilistic AI outputs to Boolean PLC con-trol signals. Validation demonstrates reliable, low-latency safety actuation with clear ar-chitectural separation between vision processing, Python-mediated policy enforcement, and PLC-based deterministic control. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

11 pages, 2066 KB  
Proceeding Paper
Vehicle Classification Using Instance Segmentation via Segmenting Objects by Locations Version 2
by Jerome Oliver S. Angeles, Paula Bianca H. Dayao and John Paul T. Cruz
Eng. Proc. 2026, 134(1), 12; https://doi.org/10.3390/engproc2026134012 - 30 Mar 2026
Viewed by 319
Abstract
Efficient vehicle classification is essential in intelligent transportation systems, contributing to improved traffic management, road safety, and automated monitoring. However, existing approaches encounter difficulties in real-world scenarios involving overlapping vehicles, varying illumination, and complex backgrounds. We developed a vehicle classification system based on [...] Read more.
Efficient vehicle classification is essential in intelligent transportation systems, contributing to improved traffic management, road safety, and automated monitoring. However, existing approaches encounter difficulties in real-world scenarios involving overlapping vehicles, varying illumination, and complex backgrounds. We developed a vehicle classification system based on the Segmenting Objects by Locations Version 2 (SOLOv2) instance segmentation framework to achieve accurate identification of seven vehicle classes: sedan, sport utility vehicle, pick-up truck, van, bus, traditional Jeepney, and modern Jeepney. The system employs a Raspberry Pi 5 integrated with an Asus Rog Eye S High Definition Webcam and a 7-inch display, providing an embedded and cost-effective platform for on-site traffic monitoring. A dataset of 2100 annotated images was developed for training and evaluation. The model achieved a mean Intersection over Union of 0.947 and a mean average precision at the intersection of union thresholds from 0.50 to 0.95 of 0.547, with strong performance for bus (a 91.05% average precision (AP)) and modern Jeepney (a 75.03% AP). These results show the technical feasibility and real-world applicability of SOLOv2 for vehicle classification, establishing a robust foundation for integrating advanced computer vision techniques into intelligent transportation systems. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1907 KB  
Proceeding Paper
Adaptive Phishing Detection and Mitigation System Using Huawei Mind Reinforcement Learning with Human Feedback
by Jesher Immanuel B. Hael, Mark Daniel S. Ortiz and Dionis A. Padilla
Eng. Proc. 2026, 134(1), 13; https://doi.org/10.3390/engproc2026134013 - 30 Mar 2026
Viewed by 339
Abstract
Phishing remains a persistent cybersecurity threat, exploiting social engineering to bypass traditional defenses. We developed a phishing detection system that integrates baseline supervised learning with Reinforcement Learning through human feedback (RLHF) to improve adaptability against evolving attack strategies. Implemented using the Huawei MindRLHF [...] Read more.
Phishing remains a persistent cybersecurity threat, exploiting social engineering to bypass traditional defenses. We developed a phishing detection system that integrates baseline supervised learning with Reinforcement Learning through human feedback (RLHF) to improve adaptability against evolving attack strategies. Implemented using the Huawei MindRLHF framework and deployed on Raspberry Pi hardware, the system was evaluated using a dataset of 135,325 email samples consisting of both phishing and legitimate messages. The baseline supervised model achieved 94.3% accuracy, while the RLHF-enhanced model, through 74 iterations, achieved improved adaptability, reaching a 96.8% accuracy with balanced precision and recall. A multi-component reward function was designed to incorporate correct classification, human agreement, confidence matching, and consistency, enabling the model to refine its decision boundaries beyond automated optimization. Real-time monitoring and feedback were facilitated through a hardware-integrated LCD interface. The results confirm enhanced detection accuracy and reduced error rates, demonstrating its viability for deployment. The findings highlight the potential of human-centered RLHF the resilience and scalability of phishing mitigation systems against emerging cyber threats. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1450 KB  
Proceeding Paper
BEMAX: A Leaf-Based Endangered Tree Classification System Using Convolutional Neural Network in Bohol Biodiversity Complex, the Philippines
by Bem Gumapac and Jocelyn Villaverde
Eng. Proc. 2026, 134(1), 14; https://doi.org/10.3390/engproc2026134014 - 30 Mar 2026
Viewed by 541
Abstract
Biodiversity monitoring in tropical ecosystems is constrained by limited infrastructure, insufficient localized datasets, and reliance on cloud-based tools. We introduce BEMAX, a lightweight convolutional neural network for offline classification of endangered tree species in the Bohol Biodiversity Complex, Philippines. A curated leaf-image dataset [...] Read more.
Biodiversity monitoring in tropical ecosystems is constrained by limited infrastructure, insufficient localized datasets, and reliance on cloud-based tools. We introduce BEMAX, a lightweight convolutional neural network for offline classification of endangered tree species in the Bohol Biodiversity Complex, Philippines. A curated leaf-image dataset from five species and an unknown class was collected using a Raspberry Pi camera. The MobileNetV2-based model achieved a 93.89% validation accuracy and an 88.33% field accuracy. Deployed on a Raspberry Pi 4 with touchscreen and camera integration, BEMAX demonstrates embedded AI as a replicable framework for conservation in data-scarce environments. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 742 KB  
Proceeding Paper
A Comparative Study of Deep Learning and Boundary-Based Registration for Thermal Imaging of Human Forearm
by Chih-Ting Shen, Ting-Ting Liu, I-Ching Kuo and Wei-Min Liu
Eng. Proc. 2026, 134(1), 15; https://doi.org/10.3390/engproc2026134015 - 30 Mar 2026
Viewed by 281
Abstract
Thermal imaging holds potential in medical diagnostics, but long-duration observational experiments, such as brachial artery occlusion, suffer from subjects’ motion, which complicates reliable temperature oscillation analysis. This work investigates the suitability of Deep Learning (DL) models for low-contrast, feature-sparse thermal registration. We examined [...] Read more.
Thermal imaging holds potential in medical diagnostics, but long-duration observational experiments, such as brachial artery occlusion, suffer from subjects’ motion, which complicates reliable temperature oscillation analysis. This work investigates the suitability of Deep Learning (DL) models for low-contrast, feature-sparse thermal registration. We examined a global–local model trained under intensity-driven unsupervised learning and a synthetic rigid supervised approach. Both models were evaluated against a rigid-body registration baseline. A comparative study was conducted to analyze the performance differences. In this specific scenario, the traditional rigid body registration achieved more reliable alignment (a Dice similarity coefficient (DSC) of 0.9760) than the DL-based approaches (DSCs of 0.9752 and 0.9746) over 2399 test images. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

11 pages, 1533 KB  
Proceeding Paper
Numerical Study and Optimization of Energy-Efficient Electro-Thermal De-Icing for Unmanned Aerial Vehicles
by Jen-Chieh Cheng and Cheng-Ying Lo
Eng. Proc. 2026, 134(1), 16; https://doi.org/10.3390/engproc2026134016 - 31 Mar 2026
Viewed by 245
Abstract
In this study, we explored an electro-thermal de-icing system that minimizes ice accumulation while optimizing energy efficiency for drones operating at various altitudes. The research introduced an ideal heating schedule for a wing de-icing system by tailoring heating cycles to different flight altitudes. [...] Read more.
In this study, we explored an electro-thermal de-icing system that minimizes ice accumulation while optimizing energy efficiency for drones operating at various altitudes. The research introduced an ideal heating schedule for a wing de-icing system by tailoring heating cycles to different flight altitudes. The heat flux is adjusted to maintain the surface temperature at 273 K, with a ±5 K margin to prevent temperature extremes. Two empirical formulas were developed to calculate the appropriate heat flux for each heating cycle, ensuring effective de-icing performance at varying altitudes. This approach eliminates the need for onboard sensors, as only the required heat flux value needs to be input to achieve effective de-icing. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 1268 KB  
Proceeding Paper
Defect Inspection of Voltage Control IC in Electric Vehicle Chargers Using Surface-Mount Technology
by Quang-Phuc Le Tran and Kuang-Chyi Lee
Eng. Proc. 2026, 134(1), 17; https://doi.org/10.3390/engproc2026134017 - 31 Mar 2026
Viewed by 265
Abstract
Ensuring the reliability of solder joints is essential for stable operation in electric vehicle chargers, particularly for components assembled using surface-mount technology. Therefore, we developed a defect inspection system for welding joint defects using a Faster Region-based Convolutional Neural Network model to classify [...] Read more.
Ensuring the reliability of solder joints is essential for stable operation in electric vehicle chargers, particularly for components assembled using surface-mount technology. Therefore, we developed a defect inspection system for welding joint defects using a Faster Region-based Convolutional Neural Network model to classify results as insufficient defect, shifting defect, and normal (pin-qualified) on voltage control IC pins. The model was trained on 72,000 pin samples and achieved a training accuracy of 99.93%. Evaluation of 65,700 pin samples resulted in an accuracy of 98.89%. The experimental results demonstrate that the system provides stable recognition of reflective solder joints, reliably identifies critical pin-level defects, and is suitable for deployment in practical inspection environments. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 4244 KB  
Proceeding Paper
Efficiency Improvement of Wireless Power Supply Track System
by Yung-Chun Wu, Chun-Cheng Su and Chieh-Lung Chang
Eng. Proc. 2026, 134(1), 18; https://doi.org/10.3390/engproc2026134018 - 30 Mar 2026
Viewed by 263
Abstract
We investigated a wireless power supply track system for automated guided vehicles. Due to the inherent limitations of wireless power transfer, its transmission efficiency is lower than that of contact-based power supply methods. To meet energy conservation and carbon reduction requirements, we proposed [...] Read more.
We investigated a wireless power supply track system for automated guided vehicles. Due to the inherent limitations of wireless power transfer, its transmission efficiency is lower than that of contact-based power supply methods. To meet energy conservation and carbon reduction requirements, we proposed methods to improve the overall system efficiency. Different from the traditional design of adding inductor or capacitor filter circuits after the rectifier circuit, this paper proposed an improved circuit structure for the pick-up end. Through theoretical analysis and discussion using two methods, valley-fill filter circuits or directly removing the filter circuits, hardware experiments have verified the feasibility of the proposed method. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

14 pages, 1839 KB  
Proceeding Paper
Digital Twin and IoT Integration for Predictive Maintenance in Civil and Structural Engineering
by Wai Yie Leong
Eng. Proc. 2026, 134(1), 19; https://doi.org/10.3390/engproc2026134019 - 31 Mar 2026
Viewed by 1372
Abstract
The growing complexity, age, and environmental exposure of civil infrastructure assets—bridges, tunnels, buildings, highways, and dams—have necessitated a transition from reactive or preventive maintenance strategies toward predictive, data-driven systems. The integration of IoT and Digital Twin (DT) technologies provides a transformative paradigm for [...] Read more.
The growing complexity, age, and environmental exposure of civil infrastructure assets—bridges, tunnels, buildings, highways, and dams—have necessitated a transition from reactive or preventive maintenance strategies toward predictive, data-driven systems. The integration of IoT and Digital Twin (DT) technologies provides a transformative paradigm for intelligent monitoring, early fault detection, and real-time lifecycle management. This paper explores the technological convergence of IoT sensor networks, edge-cloud analytics, and digital twin platforms for predictive maintenance in civil and structural engineering. The study presents a multi-layered DT–IoT integration framework designed for infrastructure assets, emphasizing interoperability, cybersecurity, and semantic data synchronization. Key research outcomes include enhanced asset availability, reduced maintenance costs, and improved safety margins. The proposed architecture incorporates sensor-level digital shadows, edge inference modules, and cloud-based analytical twins powered by hybrid machine learning and finite element models. Real-world applications and case studies from smart bridges and intelligent building systems demonstrate prediction accuracies exceeding 90% in identifying early structural fatigue indicators. Ultimately, the results underscore the strategic role of DT–IoT convergence in realizing sustainable, resilient, and self-aware civil infrastructure aligned with Industry 5.0 principles. This study provides a roadmap for digital transformation in asset management, integrating standards such as International Organization for Standardization (ISO) 23247 and ISO 19650 to ensure interoperability and lifecycle traceability. The results reinforce that predictive maintenance through DT and IoT integration is not only technically viable but essential for extending infrastructure lifespan, minimizing unplanned downtime, and achieving carbon-efficient asset operation. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

12 pages, 2073 KB  
Proceeding Paper
Binocular Stereo Vision Disparity Estimation Based on Distilled Internally Normalized Optimized Version 2 with Multi-Scale Attention Fusion
by Chang-Fu Hung, Tzu-Jung Tseng and Jian-Jiun Ding
Eng. Proc. 2026, 134(1), 20; https://doi.org/10.3390/engproc2026134020 - 31 Mar 2026
Viewed by 356
Abstract
A stereo vision framework is designed to improve disparity estimation in occluded and boundary regions, targeting autonomous driving scenarios. The proposed architecture combines frozen Distilled Internally Normalized Optimized Version 2 features with a modular three-stage attention fusion strategy, which consists of bottom-up semantic [...] Read more.
A stereo vision framework is designed to improve disparity estimation in occluded and boundary regions, targeting autonomous driving scenarios. The proposed architecture combines frozen Distilled Internally Normalized Optimized Version 2 features with a modular three-stage attention fusion strategy, which consists of bottom-up semantic propagation, top-down detail enhancement, and cross-view attention mechanisms. These stages jointly enforce semantic consistency, structural integrity, and accurate correspondence modeling. The fused features are then processed by an Iterative Geometry Encoding and Volumetric regression-based disparity estimation module for multi-stage regression and iterative refinement. A three-phase training pipeline is employed, including pretraining on SceneFlow, fine-tuning on virtual Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) benchmarks, and adaptation to the KITTI and ETH Zurich 3D benchmark dataset. The model achieves an out-of-center, non-occluded pixel error of 7.45% on KITTI2012 and a D1-all error of 4.10% on KITTI2015. Beyond quantitative performance, the proposed method produces visually superior disparity maps. The enhancements of boundary sharpness, occlusion completion, and structural coherence demonstrate the strong potential of the proposed algorithm for real-world deployment in dynamic and complex environments. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1657 KB  
Proceeding Paper
Comparative Analysis of Combustion Characteristics and Pollutant Formation in Radiant Tubes with Different Structural Designs
by Chien-Chou Lin, Tsai-Jung Chen, Chun-Chun Wang and Chien-Hsiung Tsai
Eng. Proc. 2026, 134(1), 21; https://doi.org/10.3390/engproc2026134021 - 31 Mar 2026
Viewed by 252
Abstract
This study aims to investigate the combustion characteristics, thermal distribution, and nitrogen oxide (NOx) formation of two radiant tube designs—conventional and staged combustion—under air–fuel ratios of 1:10 and 1:11. A three-dimensional numerical model was developed in ANSYS Fluent 2023 R1 to [...] Read more.
This study aims to investigate the combustion characteristics, thermal distribution, and nitrogen oxide (NOx) formation of two radiant tube designs—conventional and staged combustion—under air–fuel ratios of 1:10 and 1:11. A three-dimensional numerical model was developed in ANSYS Fluent 2023 R1 to compare flame temperature, wall temperature gradients, and pollutant emissions. The results reveal that flame temperature is the dominant factor in NOx formation. The conventional tube, with flame temperatures around 1800 °C, shows decreasing NOx emissions as the air–fuel ratio increases (corresponding to lower flame temperatures). In contrast, the staged combustion tube exhibits flame temperatures exceeding 1900 °C, where the thermal mechanism dominates, leading to a sharp increase in NOx emissions far above the conventional design. These findings highlight that in staged combustion systems, inadequate consideration of flame temperature and mixing characteristics may cause NOx control to fail or even reverse. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 1291 KB  
Proceeding Paper
Classification of Dark Condiment Sauces Through Electronic Nose Using Support Vector Machine
by Jose Julian L. Acot, Cherry Ben Jr. R. Bendol and John Paul T. Cruz
Eng. Proc. 2026, 134(1), 22; https://doi.org/10.3390/engproc2026134022 - 31 Mar 2026
Viewed by 449
Abstract
Condiment sauces such as soy sauce, fish sauce, oyster sauce, and Worcestershire sauce play a vital role in culinary practices and cultural identity, particularly in the Philippines. These sauces are distinguished by their unique volatile organic compound profiles, which define their aroma and [...] Read more.
Condiment sauces such as soy sauce, fish sauce, oyster sauce, and Worcestershire sauce play a vital role in culinary practices and cultural identity, particularly in the Philippines. These sauces are distinguished by their unique volatile organic compound profiles, which define their aroma and flavor. With the growing demand for these condiment products, there is an increasing need for accurate and efficient methods to classify them, ensuring product authenticity and strengthening quality control. However, conventional approaches such as sensory evaluation and laboratory-based chemical analysis are often expensive, time-consuming, and subjective. To address this limitation, we used an electronic nose (e-nose) system integrated with a Support Vector Machine (SVM) classifier for the classification of dark condiment sauces. The system consists of an array of MQ-series gas sensors connected to an Arduino Mega 2560 for analog-to-digital conversion, with Raspberry Pi 5 serving as the primary processing unit. Sensor data undergo preprocessing steps, including standardization and dimensionality reduction through principal component analysis, before being classified using SVM. A total of 120 samples, consisting of 40 readings per condiment type, were used for training and testing, while 60 additional samples—15 per class—were reserved for validation. The e-nose system achieved a 95% classification performance, as evaluated using a confusion matrix and overall accuracy metrics. These results demonstrate the potential of the e-nose combined with SVM as a reliable tool for condiment classification. The system offers practical applications in quality control and product authentication. Future work may extend its capabilities toward spoilage detection, the integration of different gas sensors, and the classification of a wider variety of condiment sauces. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1305 KB  
Proceeding Paper
Automatic Transliteration of Baybayin Transliterated Rizal Poems to Latin Text Using Kraken
by Johann Daniel De Guzman, Ser Alec Miguel Medrano and Dionis Padilla
Eng. Proc. 2026, 134(1), 23; https://doi.org/10.3390/engproc2026134023 - 1 Apr 2026
Viewed by 462
Abstract
We introduce an automatic transliteration system that uses Kraken optical character recognition to convert Baybayin, a Philippine national script, into its equivalent Latin script. Model training was performed using convolutional recurrent neural networks (CRNN) and multiple training epochs to optimize character and word [...] Read more.
We introduce an automatic transliteration system that uses Kraken optical character recognition to convert Baybayin, a Philippine national script, into its equivalent Latin script. Model training was performed using convolutional recurrent neural networks (CRNN) and multiple training epochs to optimize character and word recognition accuracy. The validation phase achieved a validation character accuracy of 98.1% and a validation word accuracy of 87.8%. The system’s performance further attained a 27.22% word error rate, 96.42% word detection rate, and a 75.71% Recall-Oriented Understudy for Gisting Evaluation-1 score. The results of the study indicate that the system is capable of transliterating full pages of Baybayin text. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 372 KB  
Proceeding Paper
Performance Analysis of Hammer Throwers Integrating Inertial Measurement Unit and IoT
by Li-Chun Yu and Hao-Lun Huang
Eng. Proc. 2026, 134(1), 24; https://doi.org/10.3390/engproc2026134024 - 31 Mar 2026
Viewed by 268
Abstract
Hammer throw is a complex discipline requiring strength, refined technique, and precise inter-segmental coordination. We developed an IoT-enabled system with inertial measurement units (IMUs) to provide objective, real-time analytics for coaches and athletes. IMUs were mounted on the hip, knee, and ankle to [...] Read more.
Hammer throw is a complex discipline requiring strength, refined technique, and precise inter-segmental coordination. We developed an IoT-enabled system with inertial measurement units (IMUs) to provide objective, real-time analytics for coaches and athletes. IMUs were mounted on the hip, knee, and ankle to capture tri-axial acceleration and angular velocity during the throwing action. Data were streamed wirelessly and processed to extract rotation rate profiles, joint coordination metrics, and temporal events (winds, turns, and release). Two collegiate athletes performed 10 throws, and the results were compared with video-based analysis. The IMU system captured finer-grained variations in angular velocity and acceleration during rapid rotation phases and achieved an accuracy of 93.5% in classifying higher- and lower-quality throws using cross-validated models. The system developed enables quantitative feedback and continuous progress tracking in training. The feasibility of IMU + IoT integration for hammer throw performance analysis provides a foundation for AI-assisted, on-field decision support. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 753 KB  
Proceeding Paper
Computer Vision-Based Tennis Ball Tracking Using You Only Look Once for Training Analytics
by Pei-Jung Lin, Yu-Tsen Lin, Yong-Liang Lin, Yi-Ping Lee and Shao-Wei Chang
Eng. Proc. 2026, 134(1), 25; https://doi.org/10.3390/engproc2026134025 - 2 Apr 2026
Viewed by 911
Abstract
Tennis is an exceptionally fast-paced sport where the ability to return the ball precisely to an opponent’s weak zones often determines match outcomes. Although wall practice serves as a fundamental and effective training method, accurately capturing and analyzing the spatial distribution of ball [...] Read more.
Tennis is an exceptionally fast-paced sport where the ability to return the ball precisely to an opponent’s weak zones often determines match outcomes. Although wall practice serves as a fundamental and effective training method, accurately capturing and analyzing the spatial distribution of ball impact points during high-speed rallies remains highly challenging. Leveraging computer vision, we propose a two-stage detection pipeline that integrates You Only Look Once Version 12 and MobileNetV2 to generate candidate bounding boxes, stabilized by a Kalman filter with a predict–update mechanism. This approach ensures robust and reliable object tracking, providing valuable insights into tennis training performance, placement accuracy, and actionable insights for sports analytics. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 1019 KB  
Proceeding Paper
Optimizing Compact Centrifugal Impellers for Wearable Cooling Devices: Computational Fluid Dynamics of Blade Count Effects
by Shih-Chia Wang, Fulki Shah Jahan and Dena Gabriela
Eng. Proc. 2026, 134(1), 26; https://doi.org/10.3390/engproc2026134026 - 2 Apr 2026
Viewed by 327
Abstract
Personal thermal management systems, particularly neck fans, are increasingly popular for providing hands-free, localized cooling to enhance user comfort. We investigate the aerodynamic performance of compact forward-curved centrifugal impellers (<50 mm in diameter) with 30, 28, and 24 blades, selected through benchmarking of [...] Read more.
Personal thermal management systems, particularly neck fans, are increasingly popular for providing hands-free, localized cooling to enhance user comfort. We investigate the aerodynamic performance of compact forward-curved centrifugal impellers (<50 mm in diameter) with 30, 28, and 24 blades, selected through benchmarking of commercial products. A baseline impeller and casing were reverse engineered by using 3D scanning and CAD modeling methods, followed by blade count modifications under consistent geometric constraints. CFD simulations in ANSYS Workbench 19.1 were conducted to examine velocity fields, pressure distribution, and flow rate. Results indicate that blade number significantly influences airflow and efficiency: the 28-blade impeller achieved the highest outlet velocity, while the 30-blade configuration provided smoother pressure recovery but higher flow resistance. These insights aid the design of more efficient wearable cooling devices. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 1980 KB  
Proceeding Paper
Data-Driven Long Short-Term Memory Framework for Servo System Modeling and Optimization
by Yong-Zhong Li, You-Cheng Chen, Xiang-Kai Wang and Ming-Tsung Lin
Eng. Proc. 2026, 134(1), 27; https://doi.org/10.3390/engproc2026134027 - 3 Apr 2026
Viewed by 382
Abstract
A novel data-driven modeling framework is developed for servo control using Long Short-Term Memory (LSTM) networks. The framework employs an LSTM model to directly map interpolation commands and feedback signals, such as velocity, acceleration, and jerk, to tracking errors. By adopting end-to-end architecture, [...] Read more.
A novel data-driven modeling framework is developed for servo control using Long Short-Term Memory (LSTM) networks. The framework employs an LSTM model to directly map interpolation commands and feedback signals, such as velocity, acceleration, and jerk, to tracking errors. By adopting end-to-end architecture, the method bypasses complex sequential procedures, including system identification and friction modeling, significantly reducing development time and modeling complexity. Experimental results demonstrate that the LSTM model accurately predicts servo tracking behavior and enables rapid performance evaluation and parameter optimization without performing time-consuming trajectory testing. The proposed framework offers a practical and efficient alternative to traditional model-based techniques in precision motion control. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 2493 KB  
Proceeding Paper
Campus-Scale Real-Time Waste Classification on Raspberry Pi 5 Using You Only Look Once Version 8
by Keilha Niicael L. Partosan, Euanne Jhasmine R. Villanueva and Ramon G. Garcia
Eng. Proc. 2026, 134(1), 28; https://doi.org/10.3390/engproc2026134028 - 1 Apr 2026
Viewed by 322
Abstract
A deep learning-based waste detection system was developed for campus-scale use at Mapúa University to address improper waste disposal. The system runs YOLOv8 on a Raspberry Pi 5 with a Raspberry Pi Camera v3 to detect five classes: paper, plastic, metal, food waste, [...] Read more.
A deep learning-based waste detection system was developed for campus-scale use at Mapúa University to address improper waste disposal. The system runs YOLOv8 on a Raspberry Pi 5 with a Raspberry Pi Camera v3 to detect five classes: paper, plastic, metal, food waste, and e-waste. The dataset was expanded to 3718 images at 640 × 640 through standard geometric and photometric augmentations using Roboflow. It was split into the training, testing, and validation datasets in a ratio of 80:10:10. After evaluating multiple YOLOv8 variants (n/s/m/l), YOLOv8n was chosen for its highest performance across evaluation metrics, demonstrating the best balance between accuracy and efficiency. Trained and evaluated under this setup, the model achieved an mAP50 of 0.990, a mean average precision at the intersection of union thresholds from 0.50 to 0.95 (mAP50-95) of 0.974, and precision and recall of 0.980 on the held-out validation set. Per-class mAP50 scores were 0.995 for plastic, food waste, and metal, 0.987 for e-waste, and 0.973 for paper, with the best operating point reaching an F1 score of 0.980 at a confidence level of 0.827. These results indicate that the system is suitable for campus-scale deployment, helping reduce contamination at the source and enabling earlier intervention in waste sorting. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 591 KB  
Proceeding Paper
Decomposition of Large-Scale Quadratic Unconstrained Binary Optimization Problems for Quantum Annealers and Quantum-Inspired Annealers
by Jehn-Ruey Jiang and Qiao-Yi Lin
Eng. Proc. 2026, 134(1), 29; https://doi.org/10.3390/engproc2026134029 - 7 Apr 2026
Viewed by 355
Abstract
We study the decomposition of large-scale Quadratic Unconstrained Binary Optimization Problems (QUBO) formulations for quantum and quantum-inspired annealers and propose two decomposition mechanisms. The first is one-way-one-hot (1W1H), which replaces a linear inequality with exactly one indicator bank and naturally decomposes the model [...] Read more.
We study the decomposition of large-scale Quadratic Unconstrained Binary Optimization Problems (QUBO) formulations for quantum and quantum-inspired annealers and propose two decomposition mechanisms. The first is one-way-one-hot (1W1H), which replaces a linear inequality with exactly one indicator bank and naturally decomposes the model into many small, parallel subproblems. The second is slack variable range search (SVRS), which introduces a binary-encoded slack and scans restricted windows to balance the number of subproblems and the per-subproblem variable count. Evaluation results using the P08 knapsack problem instance on the Compal Graphic Processing Unit Annealer (CGA) show that SVRS provides a favorable scalability–quality trade-off, while 1W1H remains attractive when the admissible range is small to medium and massive parallelism is available. These results motivate integrating both mechanisms into the National Central University Annealer (NCUA). Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 685 KB  
Proceeding Paper
Contactless Footprint Acquisition and Automated Identification Using Convolutional Neural Network
by Angelica A. Claros, Elmo Joaquin D. Estacion and Jocelyn F. Villaverde
Eng. Proc. 2026, 134(1), 30; https://doi.org/10.3390/engproc2026134030 - 3 Apr 2026
Viewed by 280
Abstract
Biometric systems are widely used in security and forensic applications. Conventionally, contact-based footprint scanners require physical contact, which presents significant limitations. These devices raise hygiene concerns and are impractical in field identification conditions, such as forensic investigations or disaster victim identification, where quick [...] Read more.
Biometric systems are widely used in security and forensic applications. Conventionally, contact-based footprint scanners require physical contact, which presents significant limitations. These devices raise hygiene concerns and are impractical in field identification conditions, such as forensic investigations or disaster victim identification, where quick and non-invasive methods are essential. To address these challenges, a contactless footprint acquisition and identification system was developed using image processing techniques and a Convolutional Neural Network (CNN) based on the Visual Geometry Group–16 layer architecture. The system employs a Raspberry Pi 4, a Logitech C922 camera, and a ring light to capture footprints without direct surface contact. Captured images are processed with Contrast Limited Adaptive Histogram Equalization (CLAHE) to improve contrast and mean thresholding to generate binary images for clearer feature extraction. System performance was evaluated using a multiclass confusion matrix. The CNN correctly classified 158 of 160 test images, achieving an accuracy of 98.75%. This result demonstrates higher accuracy than earlier studies that used older CNN models, such as Alex Krizhevsky’s Network and LeCun’s Network-5, which performed with fewer subjects and lower accuracy rates. The developed system shows potential for biometric security, forensic investigations, and disaster response, where contactless and reliable identification is required. Future research can expand the dataset with more diverse footprints, test performance under varied conditions, and extend the approach to other contactless biometrics such as palmprints or ears. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 892 KB  
Proceeding Paper
Applying Model Context Protocol for Offline Small Language Models in Industrial Data Management
by Nian-Ze Hu, You-Xin Lin, Hao-Lun Huang, Po-Han Lu, Chih-Chen Lin, Yu-Tzu Hung, Sing-Cih Jhang and Pei-Yu Chou
Eng. Proc. 2026, 134(1), 31; https://doi.org/10.3390/engproc2026134031 - 7 Apr 2026
Viewed by 448
Abstract
In recent years, Large Language Models (LLMs) have demonstrated strong capabilities in contextual reasoning and knowledge retrieval. However, their application in industrial domains is limited by concerns regarding data security, reliance on cloud infrastructure, and high operational costs. To address these challenges, this [...] Read more.
In recent years, Large Language Models (LLMs) have demonstrated strong capabilities in contextual reasoning and knowledge retrieval. However, their application in industrial domains is limited by concerns regarding data security, reliance on cloud infrastructure, and high operational costs. To address these challenges, this study proposes the use of the Model Context Protocol (MCP) as a middleware framework that enables the deployment of offline-operable Small Language Models (SLMs) for industrial data processing. MCP facilitates structured interaction between SLMs and external resources (e.g., databases, APIs, and processors), allowing secure and controlled data access without exposing proprietary systems. As illustrated in the proposed framework, user input is first processed by the SLM (Qwen-7B) for intent determination. When external data is required, MCP coordinates the invocation of relevant resources and integrates the returned results into the model. The SLM then generates the final response. This approach enables SLMs to perform local computation for contextual analysis and decision support while maintaining low computational requirements and full data locality. The proposed system eliminates dependence on cloud-based LLM services and enhances security and cost efficiency. Experimental results demonstrate that the MCP-based architecture provides a practical and effective solution for deploying intelligent assistants in industrial environments without relying on large-scale external AI services. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 842 KB  
Proceeding Paper
Application of Constraint Programming with Satisfiability in Nurse Scheduling
by Jehn-Ruey Jiang, Bo-Rong Chen and Wei-Hsiang Kao
Eng. Proc. 2026, 134(1), 32; https://doi.org/10.3390/engproc2026134032 - 7 Apr 2026
Viewed by 477
Abstract
We applied the Google OR-Tools Constraint Programming with Satisfiability (CP-SAT) solver to the nurse scheduling problem (NSP) to efficiently generate feasible and high-quality schedules under complex real-world constraints. The proposed model integrates hard and soft constraints, including workload balance, fairness, and staffing sufficiency, [...] Read more.
We applied the Google OR-Tools Constraint Programming with Satisfiability (CP-SAT) solver to the nurse scheduling problem (NSP) to efficiently generate feasible and high-quality schedules under complex real-world constraints. The proposed model integrates hard and soft constraints, including workload balance, fairness, and staffing sufficiency, within a unified optimization framework. A genetic algorithm (GA) is implemented as a baseline for comparison. Experimental results show that GA does not consistently produce feasible solutions, whereas CP-SAT achieves feasible schedules satisfying all constraints and is approximately 224.6 times faster than GA on the tested instance. This demonstrates CP-SAT’s superior efficiency, robustness, and applicability for solving NSP. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

13 pages, 4072 KB  
Proceeding Paper
Development of Static and Dynamic Sensor Node Energy Level Model for Different Wireless Communication Technologies
by Zoren Mabunga, Jennifer Dela Cruz and Reggie Cobarrubia Gustilo
Eng. Proc. 2026, 134(1), 33; https://doi.org/10.3390/engproc2026134033 - 8 Apr 2026
Viewed by 399
Abstract
WSN node energy forecasting contributes to improving network efficiency, extending network lifespan, and providing energy management strategies. In this study, a deep-learning-based wireless sensor network (WSN) node energy forecasting model based on Long Short-Term Memory (LSTM) and stacked-LSTM was developed across different wireless [...] Read more.
WSN node energy forecasting contributes to improving network efficiency, extending network lifespan, and providing energy management strategies. In this study, a deep-learning-based wireless sensor network (WSN) node energy forecasting model based on Long Short-Term Memory (LSTM) and stacked-LSTM was developed across different wireless communication technologies in both static and dynamic WSN setups. The performance of the deep-learning-based models was compared with traditional forecasting techniques such as Exponential Smoothing and Prophet. The results showed the superiority of LSTM and stacked-LSTM in terms of root mean square error and mean absolute error, with consistently lower values compared with the traditional forecasting techniques. The results also show that the models perform best with Long Range technology. The deep learning-based model also demonstrates its ability to perform better in both static and dynamic WSN scenarios. These results demonstrate the potential of deep-learning-based models in WSN node energy management, which can result in an optimal energy efficiency and prolong the network lifetime. Future research is needed to explore hybrid approaches to further improve the prediction performance of deep learning-based models by combining their strengths with statistical or traditional forecasting techniques. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 1667 KB  
Proceeding Paper
Cost-Effective Device with Semantic Segmentation Capability for Real-Time Detection and Classification of Marine Litter in Benthic Coastal Areas
by John Paul T. Cruz, Josiah Izaak D. Lopez, Marlon V. Maddara, Karl Justin B. Nacito, Marites B. Tabanao, Vladimer B. Kobayashi and Roben A. Juanatas
Eng. Proc. 2026, 134(1), 34; https://doi.org/10.3390/engproc2026134034 - 7 Apr 2026
Viewed by 239
Abstract
Anthropogenic marine debris (AMD) in shallow coastal benthic areas poses serious threats to ecosystems, human health, and the economy. Addressing this issue is hindered by limited data on AMD distribution and classification. We explored the use of semantic segmentation, specifically Pyramid Scene Parsing [...] Read more.
Anthropogenic marine debris (AMD) in shallow coastal benthic areas poses serious threats to ecosystems, human health, and the economy. Addressing this issue is hindered by limited data on AMD distribution and classification. We explored the use of semantic segmentation, specifically Pyramid Scene Parsing Network (PSPNet) and Deep Convolutional Neural Network for Semantic Image Segmentation, Version 3, (DeepLabV3) models, for automated AMD detection and classification. The performance was evaluated using mean intersection over union (mIoU), pixel accuracy, and frames per second (FPS). PSPNet achieved a higher mIoU (77.03%) than DeepLabV3 (75.98%), indicating better object identification. However, DeepLabV3 outperformed PSPNet in pixel accuracy (92.24% vs. 92.01%) and FPS (8.83 vs. 6.92), making it more appropriate for real-time applications. To enable real-time identification and classification of AMD, the models are deployed in a minicomputer with adequate processing power, significantly enhancing the models’ frame rate during real-time image processing. While both models are effective, DeepLabV3 is recommended for real-time AMD segmentation. The study contributes to improving AMD monitoring and management in coastal environments through AI-driven solutions. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 2189 KB  
Proceeding Paper
Automatic Packet Reporting System’s Payload Design for Development of Backup Communication System and Disaster Risk Reduction Management
by Jonald Ray M. Tadena, Marloun P. Sejera and Mark Angelo C. Purio
Eng. Proc. 2026, 134(1), 35; https://doi.org/10.3390/engproc2026134035 - 8 Apr 2026
Viewed by 292
Abstract
We developed two distinct automatic packet reporting system (APRS) payload designs to establish a reliable backup communication system for disaster risk reduction and management. The payloads are designed to perform a significant key operation, primarily APRS digital repeating (DP), enabling continuous communication access [...] Read more.
We developed two distinct automatic packet reporting system (APRS) payload designs to establish a reliable backup communication system for disaster risk reduction and management. The payloads are designed to perform a significant key operation, primarily APRS digital repeating (DP), enabling continuous communication access even in areas where conventional ground-based infrastructure is damaged by natural disasters through the relay of APRS packets to extend communication coverage. A detailed framework is designed using the standard amateur packet radio (AX.25 protocol). It specifies the structure of APRS data frames and packets, which are used to transmit alerts, emergency status updates, and text messages. This structure ensures that important information is transmitted reliably and effectively during an emergency. The designs for the APRS payloads share a common overall operating system architecture but differ in their very high frequency transceiver modules used for the amateur radio (Radiometrix BiM1H very high frequency (VHF) Narrowband Transceiver and Dorji DRA818V VHF Band Voice Transceiver Module). Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 707 KB  
Proceeding Paper
Enhancing Text-to-SPARQL Generation via In-Context Learning with Example Selection Strategies
by Eric Jui-Lin Lu and Zi-Ting Su
Eng. Proc. 2026, 134(1), 36; https://doi.org/10.3390/engproc2026134036 - 9 Apr 2026
Viewed by 394
Abstract
Large language models demonstrate strong in-context learning (ICL) capabilities, allowing them to perform diverse tasks without fine-tuning. In knowledge graph question answering (KGQA), natural language questions are translated into SPARQL queries. Existing ICL approaches mainly rely on semantic similarity, often neglecting structural features. [...] Read more.
Large language models demonstrate strong in-context learning (ICL) capabilities, allowing them to perform diverse tasks without fine-tuning. In knowledge graph question answering (KGQA), natural language questions are translated into SPARQL queries. Existing ICL approaches mainly rely on semantic similarity, often neglecting structural features. To address this limitation, we developed a structure-aware example selection strategy that integrates both semantic and structural patterns by abstracting Resource Description Framework (RDF) triples. We compare four strategies: (1) fully random, (2) semantic similarity, (3) same-type random, and (4) same-type semantic similarity. Experiments on LC-QuAD 1.0 using FLAN-T5 show that in non-fine-tuned settings, structure-aware semantic selection achieves the best results, highlighting the importance of structural congruence, while after fine-tuning, differences between strategies converge but diversity and semantic relevance remain beneficial. These findings demonstrate the critical role of example quality in ICL and provide empirical insights for KGQA design. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 2515 KB  
Proceeding Paper
Intelligent Notification Mechanism and Workflow for Legacy Programmable Logic Controller System
by Nian-Ze Hu, Po-Han Lu, Hao-Lun Huang, You-Xin Lin, Chih-Chen Lin, Yu-Tzu Hung, Sing-Cih Jhang, Pei-Yu Chou and Qi-Ren Lin
Eng. Proc. 2026, 134(1), 37; https://doi.org/10.3390/engproc2026134037 - 9 Apr 2026
Viewed by 335
Abstract
We developed a real-time alert and data management framework that integrates programmable logic controllers, RS-485 industrial communication, Structured Query Language Server, Message Queuing Telemetry Transport (MQTT), and the nodemation (n8n) automation platform, using a filling machine production line as a case study. The [...] Read more.
We developed a real-time alert and data management framework that integrates programmable logic controllers, RS-485 industrial communication, Structured Query Language Server, Message Queuing Telemetry Transport (MQTT), and the nodemation (n8n) automation platform, using a filling machine production line as a case study. The system collects and analyzes the operational status and production line data of the filling machine in real time, storing all information in a database for preservation. Through MQTT, the data is sent to n8n for automated processing. When equipment anomalies occur or data exceed predefined thresholds, the system automatically notifies maintenance personnel via communication software APIs. Additionally, users can query daily production capacity or related data using n8n’s AI functions. This architecture offers low cost, rapid deployment, cross-platform integration, and high flexibility. It not only improves anomaly handling efficiency but also preserves complete historical records, supporting trend analysis, report generation, and decision optimization, thereby assisting the filling production line in achieving long-term stable and intelligent management. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1070 KB  
Proceeding Paper
Solving the k-Hitting Set Problem with Dicke State Quantum Search
by Jehn-Ruey Jiang
Eng. Proc. 2026, 134(1), 38; https://doi.org/10.3390/engproc2026134038 - 10 Apr 2026
Viewed by 265
Abstract
An algorithm called Dicke state quantum search for the hitting set problem (DSQS-HSP), generates quantum circuits to solve the k-hitting set problem (k-HSP), by initializing the working qubits in an n-qubit Dicke state Dkn of exactly k [...] Read more.
An algorithm called Dicke state quantum search for the hitting set problem (DSQS-HSP), generates quantum circuits to solve the k-hitting set problem (k-HSP), by initializing the working qubits in an n-qubit Dicke state Dkn of exactly k qubits in 1. The quantum circuit reduces the search space size from 2n to D = nk, the number of symmetric superposition states in Dkn. A quantum-flag oracle checks the hitting condition, and a mirror-readout mechanism projects valid solutions to the output register. The circuit yields two outcome types: the all-zero string with probability (D−M)/D and solution strings, each with probability 1/D, where M is the number of solutions. The resource growth of qubits, gates, circuit depth, and circuit execution repetitions is O(nk), which remains polynomial in the best case for min(k, n − k) ≪ n/2. Experimental results using IBM Qiskit Aer Simulator confirm that the DSQS-HSP can produce quantum circuits to successfully solve the k-HSP. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 959 KB  
Proceeding Paper
Prime Number Generator Based on Chaotic System and FPGA Implementation
by Chang-Ming Wu, Yuan-Shuo Yu, Hung-Ru Lin and Chih-Hau Chang
Eng. Proc. 2026, 134(1), 39; https://doi.org/10.3390/engproc2026134039 - 9 Apr 2026
Viewed by 266
Abstract
With the growing importance of personal information security, numerous methods have been proposed for data encryption. To ensure system safety, ciphers must be unpredictable and robust. In modern Rivest–Shamir–Adleman (RSA) encryption systems, two prime numbers are required for key generation, and their randomness [...] Read more.
With the growing importance of personal information security, numerous methods have been proposed for data encryption. To ensure system safety, ciphers must be unpredictable and robust. In modern Rivest–Shamir–Adleman (RSA) encryption systems, two prime numbers are required for key generation, and their randomness and unpredictability are essential for security. In this study, we propose a secure system for generating the prime numbers used in RSA encryption. The inherent properties of chaotic systems are employed as a Pseudo Random Number Generator (PRNG), while a Ring Oscillator is utilized as a True Random Number Generator (TRNG). The Miller–Rabin algorithm is further applied to verify the primality of the numbers produced by the PRNG. The entire design is implemented on a Field Programmable Gate Array (FPGA) to achieve a fully hardware system. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 1066 KB  
Proceeding Paper
Cognitive Vision-Based Pruning Region Identification Using Deep Learning
by Monalisa S. Uysin, John Alfred Nico T. Tingson and Noel B. Linsangan
Eng. Proc. 2026, 134(1), 40; https://doi.org/10.3390/engproc2026134040 - 8 Apr 2026
Viewed by 252
Abstract
Pruning is a critical horticultural practice that requires continuous interpretation of plant structure to maintain crop health and prevent disease. Manual identification of pruning-relevant regions is labor-intensive and limits scalability in precision agriculture. This study presents a cognitive vision-based pruning region identification system [...] Read more.
Pruning is a critical horticultural practice that requires continuous interpretation of plant structure to maintain crop health and prevent disease. Manual identification of pruning-relevant regions is labor-intensive and limits scalability in precision agriculture. This study presents a cognitive vision-based pruning region identification system using a You Only Look Once version 9 model to detect lateral branches, lower leaves, and diseased leaves in Solanum lycopersicum. A custom dataset of 4905 augmented images was used for training and evaluation. The model achieved 82.86% precision, 77.24% recall, 79.96% F1-score, and 83.21% mAP. Deployment on Raspberry Pi 5 demonstrated real-time, cloud-independent edge inference, indicating the feasibility of low-cost cognitive vision systems for smart agriculture. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 1508 KB  
Proceeding Paper
Digital Microscopy-Based Barong Tagalog Textile Identification Using a Convolutional Neural Network, a Support Vector Machine, and Canny Edge Detection
by Edward C. Torralba, Jeff B. Totesora and Cyrel O. Manlises
Eng. Proc. 2026, 134(1), 41; https://doi.org/10.3390/engproc2026134041 - 13 Apr 2026
Viewed by 374
Abstract
In this study, the feasibility of using Canny edge detection for barong tagalog textile detection is examined using a digital microscope with 1000× magnification capabilities. To enhance the differentiation of textile edges, a system incorporating a digital camera was developed to apply a [...] Read more.
In this study, the feasibility of using Canny edge detection for barong tagalog textile detection is examined using a digital microscope with 1000× magnification capabilities. To enhance the differentiation of textile edges, a system incorporating a digital camera was developed to apply a Canny edge detection algorithm. Then, we evaluated the effectiveness of Canny edge detection in delineating the edges of barong tagalog textiles. The results show high accuracy in distinguishing between different textile types within the dataset. Through the adoption of Canny edge detection, the accurate identification of jusi, piña-silk, and cocoon-silk textiles was enhanced. The model demonstrated a 90% accuracy rate (27 out of 30 trials) in detecting jusi, piña-silk, and cocoon-silk textiles based on over 625 data points. The model accurately classified the textiles. The limitations of the developed model can be addressed by expanding the dataset and including a broader range of barong tagalog textiles, thus enhancing the model’s precision and applicability. The developed model in this study contributes to creating a valuable dataset for barong tagalog textiles, and the system’s potential for real-time applications in local stores, manufacturers, and research facilities can be envisioned. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 450 KB  
Proceeding Paper
Class Entity Identification Based on Large Language Models: A Choice Between Classification and Generation
by Eric Jui-Lin Lu and Cheng-Hao Yang
Eng. Proc. 2026, 134(1), 42; https://doi.org/10.3390/engproc2026134042 - 10 Apr 2026
Viewed by 283
Abstract
Large language models (LLMs) have been widely applied to knowledge graph question answering (KGQA) systems. Recent Text-to-SPARQL studies have demonstrated that generation performance can achieve an F1 score exceeding 90%. Further error analysis has categorized common errors into entity translation errors, entity position [...] Read more.
Large language models (LLMs) have been widely applied to knowledge graph question answering (KGQA) systems. Recent Text-to-SPARQL studies have demonstrated that generation performance can achieve an F1 score exceeding 90%. Further error analysis has categorized common errors into entity translation errors, entity position errors, and resource description framework (RDF) triple-count errors, with the latter accounting for 24% of all errors. Notably, nearly 90% of RDF triple-count errors occur when the triples involve class entities. Previous research has shown that incorporating prompts can effectively enhance model performance. Based on the results, we predicted whether a question contains a class entity and the number of RDF triples in the corresponding query to reduce RDF triple-count errors in large language models by providing precise task-related information through prompt design. Since both strategies are classification-oriented, two implementation paradigms were established: traditional classification architectures and generative modeling. They were compared in terms of performance. For classification-based architectures, we employed Bidirectional Encoder Representations from Transformers (BERT) and the Robustly Optimized BERT Approach (RoBERTa) to obtain question embeddings for classification. For the generative approach, we adopted the Instruction-Tuned Text-to-Text Transfer Transformer (Flan-T5). Experimental results show that the generative model slightly outperforms conventional classification architectures, indicating that generative approaches can achieve higher prediction accuracy and provide more reliable information without the need for additional complex encoder designs, thereby improving the overall quality of Text-to-SPARQL generation. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

15 pages, 3825 KB  
Proceeding Paper
Development of an Augmented Sungka Board Using Fuzzy Logic and Heuristic Search
by Albert Dylan David, Raymund Sean Clapano and Analyn Yumang
Eng. Proc. 2026, 134(1), 43; https://doi.org/10.3390/engproc2026134043 - 10 Apr 2026
Viewed by 355
Abstract
We developed an augmented Sungka board that integrates traditional Filipino gameplay with embedded sensor technology. Each pit is equipped with load cell sensors and HX711 analog-to-digital converters to accurately detect marble distribution and movement in real time. A Raspberry Pi 4 serves as [...] Read more.
We developed an augmented Sungka board that integrates traditional Filipino gameplay with embedded sensor technology. Each pit is equipped with load cell sensors and HX711 analog-to-digital converters to accurately detect marble distribution and movement in real time. A Raspberry Pi 4 serves as the central controller, handling sensor data acquisition, game state processing, rule enforcement, and output display through a liquid crystal display. The system enables automatic score tracking, move validation, and real-time board updates without altering the physical structure or rules of Sungka. A rule-based decision algorithm using fuzzy logic and heuristic search evaluates possible moves in constant time, allowing seamless real-time interaction. Across 10,000 simulated games, the algorithm achieved win rates of 84.9% against random, 77.7% against greedy, and 56.3% against exact-match strategies, with statistically consistent performance. By combining reliable hardware sensing with intelligent decision support, the proposed system enhances engagement while preserving the cultural authenticity of Sungka. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1325 KB  
Proceeding Paper
Determining the Freshness of Milkfish (Chanos chanos) Using Electronic Nose
by John Paulo D. Fernandez, Juhyoung Lee and Meo Vincent C. Caya
Eng. Proc. 2026, 134(1), 44; https://doi.org/10.3390/engproc2026134044 - 13 Apr 2026
Viewed by 368
Abstract
Milkfish (Chanos chanos), a widely consumed fish in the Philippines, is highly perishable, and conventional freshness assessments based on physical and olfactory inspection are often subjective and unreliable. To address this, we introduce an electronic nose system for the accurate classification [...] Read more.
Milkfish (Chanos chanos), a widely consumed fish in the Philippines, is highly perishable, and conventional freshness assessments based on physical and olfactory inspection are often subjective and unreliable. To address this, we introduce an electronic nose system for the accurate classification of milkfish freshness based on spoilage-related gas emissions, namely methane, ammonia, hydrogen sulfide, and trimethylamine. The system integrates the MQ-series sensors and Taguchi gas sensor with Arduino Nano and Raspberry Pi 5 for data acquisition and signal processing. The k-nearest neighbor algorithm was used for classification, and its performance was evaluated using a confusion matrix. The data was gathered from 100 samples, consisting of 50 fresh and 50 spoiled fish. The evaluation demonstrated a peak classification accuracy of 92% for k-values between 1 and 15, confirming the system’s reliability. These findings indicate the system’s potential as a practical, low-cost, and efficient tool for enhancing consumer safety and quality assurance in the fish supply chain. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1597 KB  
Proceeding Paper
Deep Learning-Based Identification of Invasive Aquatic Plant Species Using Residual Network-50
by Josh Reyes, Jacob Velasco and Jocelyn Villaverde
Eng. Proc. 2026, 134(1), 45; https://doi.org/10.3390/engproc2026134045 - 13 Apr 2026
Viewed by 257
Abstract
CNN with a Residual Network-50 (ResNet-50) architecture installed on a Raspberry Pi 5 is used to detect invasive aquatic plants in this study. By using object detection, the model recognizes water hyacinth, water lettuce, and water thyme and labels and bounds them accordingly. [...] Read more.
CNN with a Residual Network-50 (ResNet-50) architecture installed on a Raspberry Pi 5 is used to detect invasive aquatic plants in this study. By using object detection, the model recognizes water hyacinth, water lettuce, and water thyme and labels and bounds them accordingly. Images are taken by hand or at predetermined times, and verified detections are saved for later use. The adjusted ResNet-50 demonstrated 80.1% precision and 44.35% recall on validation, with 86.78% validation accuracy and 86.08% test accuracy. Target species from actual samples were successfully identified by the system. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1728 KB  
Proceeding Paper
Hardware-in-the-Loop Simulation of a Controller Area Network-Based Battery Management System for Electric-Powered Emergency Response Boats
by Lorenzo S. Decena, Jozef Marie A. Gutierrez and Febus Reidj G. Cruz
Eng. Proc. 2026, 134(1), 46; https://doi.org/10.3390/engproc2026134046 - 13 Apr 2026
Viewed by 399
Abstract
We developed a hardware-in-the-loop simulation of a battery management system (BMS) using controller area network (CAN) as the communication backbone for electric-powered response boats in flood rescue. A LiFePO4 pack and discharge motor/charger were modeled in MATLAB/Simulink/Simscape, while an STM32 Nucleo-F446RE executed CAN [...] Read more.
We developed a hardware-in-the-loop simulation of a battery management system (BMS) using controller area network (CAN) as the communication backbone for electric-powered response boats in flood rescue. A LiFePO4 pack and discharge motor/charger were modeled in MATLAB/Simulink/Simscape, while an STM32 Nucleo-F446RE executed CAN messaging. The BMS monitored voltage, current, temperature, and state of charge. Results indicate CAN’s reliability under rescue-like disturbances: priority arbitration delivered over-temperature and over-current warnings ahead of routine telemetry; error detection and retransmission preserved data integrity; and bus-load analysis showed low latency for urgent frames without interrupting state-of-charge reporting, improving situational awareness and reducing operator risk. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 1166 KB  
Proceeding Paper
Development of Transactional Filipino Sign Language Recognition System Using MediaPipe and Gated Recurrent Units
by Angela Cardano, Franz Railey Columna and Jocelyn Villaverde
Eng. Proc. 2026, 134(1), 47; https://doi.org/10.3390/engproc2026134047 - 14 Apr 2026
Viewed by 369
Abstract
Persistent communication barriers for the deaf and hard-of-hearing community in the Philippines are addressed in this study by developing a Filipino Sign Language Recognition (SLR) system. The system focuses on transactional signs commonly used in commercial environments such as markets and public facilities, [...] Read more.
Persistent communication barriers for the deaf and hard-of-hearing community in the Philippines are addressed in this study by developing a Filipino Sign Language Recognition (SLR) system. The system focuses on transactional signs commonly used in commercial environments such as markets and public facilities, thereby filling a gap left by existing SLR models. A vision-based approach was adopted, employing MediaPipe for landmark detection and Gated Recurrent Units for translating signs into text. To train the model, a custom dataset comprising 1065 video samples of 26 transactional signs was created, accounting for subtle variations in individual signing styles. The complete system was implemented on a Raspberry Pi 5 equipped with a webcam and touchscreen display. When evaluated on unseen data, the system achieved a recognition accuracy of 87%, demonstrating its potential for real-world applications in supporting commercial interactions for deaf and hard-of-hearing individuals. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

13 pages, 1407 KB  
Proceeding Paper
Enhanced Sensor-Based Automatic Fire Suppression System for Residential Kitchen Safety
by Chimie Blanche G. Cangco, Marq Ryan A. Hernandez and Joseph Bryan G. Ibarra
Eng. Proc. 2026, 134(1), 48; https://doi.org/10.3390/engproc2026134048 - 14 Apr 2026
Viewed by 536
Abstract
Fire outbreaks, whether caused naturally or unintentionally, pose serious threats to safety, especially in household environments such as kitchens. Common triggers include overheated personal devices, electrical malfunctions, and unattended cooking appliances. This study aims to develop and enhance an automated fire suppression system [...] Read more.
Fire outbreaks, whether caused naturally or unintentionally, pose serious threats to safety, especially in household environments such as kitchens. Common triggers include overheated personal devices, electrical malfunctions, and unattended cooking appliances. This study aims to develop and enhance an automated fire suppression system designed specifically for residential kitchen settings. The system integrates multiple sensors, photoelectric, ionization, and flame detectors, paired with an Arduino microcontroller to ensure accurate detection and timely activation of a servo mechanism that triggers either a Class A or Class K fire extinguisher. Through controlled testing using both solid and liquid combustible materials, we examined key variables, including sensor placement, height, and nozzle angle. The results from 15 trials per session revealed a correlation coefficient exceeding 0.90 between detection time and distance and the significance level of an analysis of variance of less than 0.05, indicating that increased distance significantly affects response time. The percent error remained below 6.7% across all tests, with strong correlations above 0.8 between combustible material type and the corresponding extinguisher class. This research contributes to the advancement of intelligent fire suppression systems by enhancing detection accuracy, reducing false triggers, and optimizing efficient sensor configurations for residential safety. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 964 KB  
Proceeding Paper
Determination of Animal-Based and Plant-Based Meat Products with an Electronic Nose Using a Fuzzy Logic Algorithm
by Kyla Marie W. Calalang, Vince Samuel R. De Peña and Jocelyn F. Villaverde
Eng. Proc. 2026, 134(1), 49; https://doi.org/10.3390/engproc2026134049 - 13 Apr 2026
Viewed by 236
Abstract
The increasing global demand for plant-based meat alternatives, driven by concerns for environmental sustainability, animal welfare, and health, has led to a growing need for reliable food authentication methods. Animal-based and plant-based meat products are visually similar, which poses a challenge for consumers [...] Read more.
The increasing global demand for plant-based meat alternatives, driven by concerns for environmental sustainability, animal welfare, and health, has led to a growing need for reliable food authentication methods. Animal-based and plant-based meat products are visually similar, which poses a challenge for consumers to distinguish them. We developed an electronic nose (e-nose) system with an array of MQ gas sensors (MQ-2, MQ-3, MQ-7, MQ-135, MQ-136, MQ-138), an Arduino MEGA microcontroller, and an LCD for displaying results. A fuzzy logic algorithm was implemented to process sensor data and enable decision-making through membership functions and IF-THEN rule evaluation to classify meat products as either animal meat or plant-based meat. The system performance was validated with 20 independent test samples. Determination accuracy for both categories, as well as the overall accuracy, was assessed using a confusion matrix. The findings demonstrate that the e-nose system can reliably distinguish between animal-based and plant-based meat products. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

12 pages, 796 KB  
Proceeding Paper
Design of a Lightweight Video-Based Ear Biometric System on Raspberry Pi 5 Using You Only Look Once Version 12 and EfficientNet-4
by Kristian Emmanuel Padilla, Michael Robin Saculsan and John Paul Cruz
Eng. Proc. 2026, 134(1), 50; https://doi.org/10.3390/engproc2026134050 - 14 Apr 2026
Viewed by 445
Abstract
Recent advances in ear biometrics have yielded increasingly accurate detection and recognition methods, driven by the ear’s uniqueness and permanence as a non-invasive biometric modality. Nonetheless, several limitations persist, including computationally demanding models, inconsistent evaluation metrics, and portable systems restricted by manual capture [...] Read more.
Recent advances in ear biometrics have yielded increasingly accurate detection and recognition methods, driven by the ear’s uniqueness and permanence as a non-invasive biometric modality. Nonetheless, several limitations persist, including computationally demanding models, inconsistent evaluation metrics, and portable systems restricted by manual capture and limited datasets. To address these challenges, we developed a lightweight, video-based ear biometric system implemented on the Raspberry Pi 5. The system integrates You Only Look Once Version 12 (YOLOv12) for ear detection, EfficientNet-4 for feature extraction, and k-Nearest Neighbors (k-NNs) for recognition. Its robust hardware platform combines Raspberry Pi 5 with the Raspberry Pi AI Camera and AI HAT+. To train, fine-tune, and optimize YOLOv12 and EfficientNet-4, we used the Visual Geometry Group (VGG)Face-Ear dataset for training and the Unconstrained Ear Recognition Challenge 2019 dataset for validation, with k-NN employed for classification. The system is evaluated for classification accuracy and system-level performance. 13 participants, comprising 10 enrolled and three unenrolled subjects, participated in testing the system. The enrolled participants registered in the system were correctly identified, whereas unenrolled participants were excluded and rejected. The system achieved 92.31% accuracy, 95.45% precision, 96.97% recall, and an F1-score of 0.95, confirming the feasibility of deploying advanced ear biometric methods on embedded, resource-constrained devices. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 1805 KB  
Proceeding Paper
Fusarium Disease Identification in Pineapple Using Convolutional Neural Network with False-Prediction Interpretability via Local Interpretable Model-Agnostic Explanations
by King Arjei Briol, Robbie Rick Gutierrez and Rosemarie Pellegrino
Eng. Proc. 2026, 134(1), 51; https://doi.org/10.3390/engproc2026134051 - 15 Apr 2026
Viewed by 277
Abstract
We developed an automated system for detecting Fusarium infection in pineapple leaves and fruits using a two-input onvolutional neural network. Implemented on a Raspberry Pi 5 with a high-quality camera, the system analyzes image pairs, fruit, and leaves of healthy and infected samples. [...] Read more.
We developed an automated system for detecting Fusarium infection in pineapple leaves and fruits using a two-input onvolutional neural network. Implemented on a Raspberry Pi 5 with a high-quality camera, the system analyzes image pairs, fruit, and leaves of healthy and infected samples. To build the dataset, pineapple images were inoculated with Fusarium, photographed daily for 15 days, then augmented through geometric and color transformations, producing 1500 image pairs. Model transparency was enhanced by using local interpretable model-agnostic explanations (LIME). Evaluated with a confusion matrix, the model achieved an 89.61% accuracy using 77 infected and 77 non-infected image pairs for testing. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1155 KB  
Proceeding Paper
Electronic Nose-Based Classification of Honey Brands Using Extreme Gradient-Boosted Decision Tree
by Mark Jasper R. Iglesias, Xandre Adrian M. Nicolas and Meo Vincent C. Caya
Eng. Proc. 2026, 134(1), 52; https://doi.org/10.3390/engproc2026134052 - 15 Apr 2026
Viewed by 302
Abstract
Honey is one of the most valued natural food products, yet it remains highly vulnerable to fraud through mislabeling and adulteration, practices that mislead consumers and compromise food safety. We develop a low-cost and portable electronic nose (e-nose) system for classifying locally available [...] Read more.
Honey is one of the most valued natural food products, yet it remains highly vulnerable to fraud through mislabeling and adulteration, practices that mislead consumers and compromise food safety. We develop a low-cost and portable electronic nose (e-nose) system for classifying locally available honey brands in the Philippines. The system integrates an array of eight MQ gas sensors to detect volatile organic compounds (VOCs), with an Arduino Mega 2560 handling data acquisition and a Raspberry Pi 5 executing data processing and classification. An Extreme Gradient-Boosted Decision Tree (XGBoost) algorithm was applied to analyze the VOC profiles of three honey brands, each with 38 samples, resulting in a total dataset of 114 samples. The dataset was divided into training, testing, and validation sets to assess the system’s classifying and predictive performance, with accuracy evaluated using a 3 × 3 confusion matrix. The results showed that the system effectively distinguished between honey brands, achieving a validation accuracy of 87.50%, corresponding to 21 out of 24 correctly identified validation trials. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1495 KB  
Proceeding Paper
Defect Identification of Trinitario Cacao Beans Using Residual Network-50 for Quality Control
by Jed Nathan L. Villapando, Kyle Aldrich R. Bordonada and Glenn V. Magwili
Eng. Proc. 2026, 134(1), 53; https://doi.org/10.3390/engproc2026134053 - 13 Apr 2026
Viewed by 248
Abstract
Cacao grading in the Philippines has relied on slow and inconsistent visual inspection. To effectively detect defects in Trinitario cacao beans, we developed a compact, low-cost computer vision system using single-bean images captured with a Raspberry Pi 5 and Camera Module 3 under [...] Read more.
Cacao grading in the Philippines has relied on slow and inconsistent visual inspection. To effectively detect defects in Trinitario cacao beans, we developed a compact, low-cost computer vision system using single-bean images captured with a Raspberry Pi 5 and Camera Module 3 under controlled lighting and distance conditions. The dataset comprises 1565 images, partitioned into training (80%), validation (10%), and testing (10%) sets. Each image was resized to 224 × 224 pixels, normalized with ImageNet statistics, and subjected to light augmentation. A ResNet-50 model was fine-tuned through transfer learning, employing AdamW optimization, warmup–cosine scheduling, label smoothing, exponential moving average, and early stopping, to classify beans into five categories: good, moldy, slaty, germinated, and over-fermented. On the held-out test set, the model achieved a 94.0% accuracy, strong per-class F1 scores, and high one-vs-rest mean average precision. Compared with a Visual Geometry Group-16 approach, which attained a 90.67% accuracy, the developed system improved performance by 3.3% while remaining inexpensive and easy to deploy. The lightweight system provides reliable and scalable cacao bean screening. Further improvements are anticipated through the expansion of underrepresented classes and refinement of class-specific thresholds. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 623 KB  
Proceeding Paper
Neocaridina (Cherry Shrimp) Sex Identification Using You Only Look Once Version 9
by Joshua Rei Y. Abundo, Jesus Raphael C. Aquino and Jocelyn F. Villaverde
Eng. Proc. 2026, 134(1), 54; https://doi.org/10.3390/engproc2026134054 - 16 Apr 2026
Viewed by 555
Abstract
Cherry shrimp (Neocaridina davidi) are a popular ornamental freshwater species known for their bright colors and ability to thrive in a variety of tank environments. However, due to their small size and the subtle differences between males and females, it can [...] Read more.
Cherry shrimp (Neocaridina davidi) are a popular ornamental freshwater species known for their bright colors and ability to thrive in a variety of tank environments. However, due to their small size and the subtle differences between males and females, it can be challenging to determine their sex. A You Only Look Once Version 9 (YOLOv9) object identification model and a Raspberry Pi 4-based system are used in this study to infer and classify the sex of cherry shrimp. A graphical user interface facilitates image collection and classification and displays the results. We developed a Raspberry Pi 4-based device with a camera module that captures images of cherry shrimp and integrated a DynamicDet architecture with Programmable Gradient Information and Generalized Efficient Layer Aggregation Networks to classify the sex of cherry shrimp. We evaluated the performance of the model using a confusion matrix to measure the accuracy of the sex classification. A confusion matrix was used to assess the collected data, and the system achieved an accuracy of 85.00%. The researchers suggest expanding the dataset to include more color variations, focusing on adding more robust male shrimp datasets, enabling the device to function without an enclosure, and updating the technology for faster inference. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 2032 KB  
Proceeding Paper
Tagalog Lip-Reading System Using 3D Convolutional Neural Network with Bidirectional Long Short-Term Memory
by Azer David V. Pascual, Titus Joaquin G. Ayo and Charmaine C. Paglinawan
Eng. Proc. 2026, 134(1), 55; https://doi.org/10.3390/engproc2026134055 - 16 Apr 2026
Viewed by 395
Abstract
We present a Tagalog lip-reading system designed to enhance communication accessibility for individuals with hearing impairments. Existing lip-reading models focus on English and other major languages and cannot recognize Tagalog visual speech patterns. To address this gap, we implemented 3D Convolutional Neural Network [...] Read more.
We present a Tagalog lip-reading system designed to enhance communication accessibility for individuals with hearing impairments. Existing lip-reading models focus on English and other major languages and cannot recognize Tagalog visual speech patterns. To address this gap, we implemented 3D Convolutional Neural Network combined with Bidirectional Long Short-Term Memory network, supported by a custom Tagalog dataset of common words. This architecture achieved an average character error rate of 10.09%, word error rate of 24.08%, and overall word accuracy of 76.27%, demonstrating promising recognition accuracy for Tagalog lip movements. By introducing the Tagalog-specific lip-reading framework, the potential of deep learning-based visual speech recognition was validated to support inclusive technologies, with applications in daily communication, education, and assistive tools for the Filipino deaf community. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 1313 KB  
Proceeding Paper
Detection and Segmentation of Surface Corrosion in Steel-Based Hand Tools Using You Only Look Once Version 8
by Allen Gabriel B. Tolentino, Vynz Alfred B. Raynes and Analyn N. Yumang
Eng. Proc. 2026, 134(1), 56; https://doi.org/10.3390/engproc2026134056 - 13 Apr 2026
Viewed by 194
Abstract
A YOLOv8-based instance segmentation model is developed for detecting and segmenting surface corrosion in steel hand tools in this study. The model was trained using a 7000-image custom dataset and deployed on a Raspberry Pi 5-based setup with a controlled lighting environment. The [...] Read more.
A YOLOv8-based instance segmentation model is developed for detecting and segmenting surface corrosion in steel hand tools in this study. The model was trained using a 7000-image custom dataset and deployed on a Raspberry Pi 5-based setup with a controlled lighting environment. The testing results showed an overall segmentation precision of 86.62%, a recall of 82.90%, and an F1-score 84.72%. Corrosion segmentation struggles with a precision of 75.02%, a recall of 17.40%, an F1-score of 28.25% and a Dice coefficient of 72.21%, demonstrating effective tool detection and classification, but it struggles on small corrosion patches, emphasizing the need for architectural and dataset enhancements. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 3650 KB  
Proceeding Paper
The Effect of Focal Length Variations on Convolutional Neural Network-Based Fabric Classifications
by Jhamil Gutierrez and Jocelyn Villaverde
Eng. Proc. 2026, 134(1), 57; https://doi.org/10.3390/engproc2026134057 - 16 Apr 2026
Viewed by 292
Abstract
This study investigated the impact of image capture distance on the performance of convolutional neural networks (CNNs) in classifying fabrics. Unlike previous works that rely solely on digital zoom and data augmentation to simulate multi-scale variations, this research explores the use of physically [...] Read more.
This study investigated the impact of image capture distance on the performance of convolutional neural networks (CNNs) in classifying fabrics. Unlike previous works that rely solely on digital zoom and data augmentation to simulate multi-scale variations, this research explores the use of physically captured images at far, mid-range, and near focal lengths using a camera with an attached varifocal lens. Fabric samples from three categories of Cotton, Linen, and Silk were imaged under consistent lighting to create an image dataset with a total of 1350 images used to train CNN models via transfer learning, with MobileNetV2 and ResNet50 as the baseline architectures. Classification performance was evaluated separately on each focal subset and on their combined dataset to test the trained model generalization capability. Results showed an absolute accuracy gain of 20.57% with MobileNetV2 and 9.78% for ResNet50 while performing with an improved accuracy at 98.42% for MobileNetV2 and ResNet50 at 96.30% Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 1047 KB  
Proceeding Paper
Image Colorization of Fruits and Vegetables Using Convolutional Kolmogorov–Arnold Networks
by Mico Kent P. Malatag, Jhanna D. Vicente and John Paul T. Cruz
Eng. Proc. 2026, 134(1), 58; https://doi.org/10.3390/engproc2026134058 - 16 Apr 2026
Viewed by 216
Abstract
Image colorization transforms monochrome images into full-colored versions, which improves image restoration in fields such as art, history, and medicine. AI models, such as convolutional neural networks and generative adversarial networks, are widely used, but they have limitations in generalization and interpretability. Therefore, [...] Read more.
Image colorization transforms monochrome images into full-colored versions, which improves image restoration in fields such as art, history, and medicine. AI models, such as convolutional neural networks and generative adversarial networks, are widely used, but they have limitations in generalization and interpretability. Therefore, we applied the Convolutional Kolmogorov–Arnold Network (CKAN), a new neural architecture that adds a convolutional layer to the Kolmogorov–Arnold Network for colorizing grayscale images of fruits and vegetables. A dataset of different varieties of fruits and vegetables was used, and the model’s performance was evaluated using the structural similarity index (SSIM) and mean squared error (MSE). After testing the model, the results showed that the CKAN colorized images achieved the desired outcome, consistently having a high SSIM score (up to 0.9) and a low MSE score (<100.0). This confirms CKAN’s potential for effective image colorization and highlights its possible applications in other computer vision tasks. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 1809 KB  
Proceeding Paper
Real-Time Classification of Guinea Pig Using You Only Look Once Version 9-Small and Raspberry Pi 5
by Jethro Ray P. Antiojo, John Patrick B. Bonilla and John Paul T. Cruz
Eng. Proc. 2026, 134(1), 59; https://doi.org/10.3390/engproc2026134059 - 17 Apr 2026
Viewed by 287
Abstract
We developed a real-time guinea pig breed classification system using You Only Look Once Version 9 (YOLOv9)-small, deployed on a Raspberry Pi 5 with Camera Module 3 and Hailo-8L acceleration module. The system targeted three breeds, Abyssinian, American, and Peruvian, using a dataset [...] Read more.
We developed a real-time guinea pig breed classification system using You Only Look Once Version 9 (YOLOv9)-small, deployed on a Raspberry Pi 5 with Camera Module 3 and Hailo-8L acceleration module. The system targeted three breeds, Abyssinian, American, and Peruvian, using a dataset of 4500 images split into a 70:20:10 ratio for training, validation, and testing. After optimization for Hailo-8L, the model was tested on live samples, with hamsters included as an unknown class. A total of 600 frame blocks were extracted from the video input and analyzed using a multi-class confusion matrix. Results showed an 89% overall accuracy (94.67% for Abyssinian, 94.33% for American, 98.67% for Peruvian, and 90.33% for unknown classification accuracy). The results showed the feasibility of deploying YOLOv9-small on embedded devices for accurate and real-time animal classification. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 2334 KB  
Proceeding Paper
Automated IoT-Based Water Quality Monitoring and Control with Fuzzy Logic for Intensive Aquaculture of Oreochromis niloticus
by Andree Scepter Guansing, Adrian Nallatan and Glenn Magwili
Eng. Proc. 2026, 134(1), 60; https://doi.org/10.3390/engproc2026134060 - 16 Apr 2026
Viewed by 441
Abstract
The Bureau of Fisheries and Aquatic Resources Tilapia Industry Roadmap (2022–2025) emphasizes the need for technological innovation in Philippine aquaculture. We developed an automated IoT-based monitoring and control system for Oreochromis niloticus using fuzzy logic for the dynamic regulation of temperature, dissolved oxygen, [...] Read more.
The Bureau of Fisheries and Aquatic Resources Tilapia Industry Roadmap (2022–2025) emphasizes the need for technological innovation in Philippine aquaculture. We developed an automated IoT-based monitoring and control system for Oreochromis niloticus using fuzzy logic for the dynamic regulation of temperature, dissolved oxygen, pH, ammonia, total dissolved solids, and turbidity. The system integrates sensors and a web-based interface for real-time data access and management of aeration, filtration, and temperature. Experimental results show the improved stability of water quality, reduced fish mortality, and enhanced growth performance compared with conventional setups. The system demonstrates a practical and sustainable approach to intensifying tilapia aquaculture through smart automation. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 1056 KB  
Proceeding Paper
Low-Resolution Script Recognition for Chinese Characters with Similar Radicals Based on Local Relations and Global Statistical Features
by Yu-Jie Yao, Pin-Wen Huang and Jian-Jiun Ding
Eng. Proc. 2026, 134(1), 61; https://doi.org/10.3390/engproc2026134061 - 17 Apr 2026
Viewed by 197
Abstract
Chinese handwritten character recognition is challenging due to structural similarities among visually similar radicals and limited available training data, especially in the low-resolution case. In this study, a multi-dimensional feature fusion method combining a histogram of oriented gradients, Hu moments, Zernike moments, local [...] Read more.
Chinese handwritten character recognition is challenging due to structural similarities among visually similar radicals and limited available training data, especially in the low-resolution case. In this study, a multi-dimensional feature fusion method combining a histogram of oriented gradients, Hu moments, Zernike moments, local binary patterns, gray-level co-occurrence matrix, and stroke-based descriptors is proposed. Region-specific segmentation strategies enable fine-grained feature extraction, and recursive feature elimination with cross-validation effectively reduces feature redundancy. Experiments demonstrate that the proposed algorithm has superior recognition performance compared to existing methods, including deep learning-based methods, especially under data-constrained or low-resolution scenarios, highlighting the effectiveness and practicality of the proposed approach. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 728 KB  
Proceeding Paper
Portable Image Classification System for Identifying Banana Leaf Diseases and Severity
by Angelica L. Genove, Aaron Cedric C. Nufable and Glenn V. Magwili
Eng. Proc. 2026, 134(1), 62; https://doi.org/10.3390/engproc2026134062 - 17 Apr 2026
Viewed by 280
Abstract
Banana production is a vital agricultural sector in the Philippines and faces major threats from Bacterial Wilt, Banana Bunchy Top Disease, Sigatoka, and Panama. We developed a portable, non-destructive detection system using image processing and deep learning to classify banana leaf diseases. Using [...] Read more.
Banana production is a vital agricultural sector in the Philippines and faces major threats from Bacterial Wilt, Banana Bunchy Top Disease, Sigatoka, and Panama. We developed a portable, non-destructive detection system using image processing and deep learning to classify banana leaf diseases. Using the MobileNetV2 architecture, the system achieved 74.6% accuracy, with the highest performance on Bacterial Wilt (F1 = 0.76) and Healthy leaves (F1 = 0.85), and lower results on Banana Bunchy Top Disease (F1 = 0.50). The system provided severity scoring through Open Source Computer Vision Library segmentation: low (0–20%), moderate (21–40%), and high (>41%). Despite power and thermal constraints, the prototype proved effective for early, field-ready disease diagnosis. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1321 KB  
Proceeding Paper
Sandstorm Image Reconstruction by Adaptive Prior, Selective Enhancement, and Sky Detection
by Hsiao-Chu Huang, Tzu-Jung Tseng and Jian-Jiun Ding
Eng. Proc. 2026, 134(1), 63; https://doi.org/10.3390/engproc2026134063 - 21 Apr 2026
Viewed by 154
Abstract
In sandstorm environments, a large number of suspended particles in the air absorb and scatter light, causing strong color bias, low contrast, and blurred details in images. These degradations reduce the reliability of computer vision applications in surveillance systems, intelligent transportation systems, unmanned [...] Read more.
In sandstorm environments, a large number of suspended particles in the air absorb and scatter light, causing strong color bias, low contrast, and blurred details in images. These degradations reduce the reliability of computer vision applications in surveillance systems, intelligent transportation systems, unmanned aerial vehicle monitoring, and outdoor autonomous driving systems. A complete sandstorm image enhancement method was developed in this study by combining sky detection, color correction, contrast enhancement, and adaptive dark channel prior (ADCP) dehazing. The Lab color space was used to correct the color bias. The L channel was enhanced using normalized gamma correction and contrast-limited adaptive histogram equalization to improve brightness and contrast. Then, the sky region is detected to avoid over-processing, preserving the natural appearance of the sky region. Finally, ADCP is applied to non-sky regions for further dehazing. Experiments show that the proposed method provides better subjective and objective performance compared to other algorithms. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 3391 KB  
Proceeding Paper
Self-Coupled Optical Waveguide-Based Tunable Photonic Structure for Spectral Control and Transmission Response Simulation
by Charmaine C. Paglinawan, Arnold C. Paglinawan, Benjamin B. Dingel and Gwen G. Evangelista
Eng. Proc. 2026, 134(1), 64; https://doi.org/10.3390/engproc2026134064 - 21 Apr 2026
Viewed by 175
Abstract
We propose a novel self-coupled optical waveguide (SCOW+) architecture that enhances spectral control in integrated photonic circuits. Derived from the foundational SCOW platform, SCOW+ introduces a tunable ring resonator coupled with an all-pass filter to achieve sharp, periodic transmission dips with adjustable free [...] Read more.
We propose a novel self-coupled optical waveguide (SCOW+) architecture that enhances spectral control in integrated photonic circuits. Derived from the foundational SCOW platform, SCOW+ introduces a tunable ring resonator coupled with an all-pass filter to achieve sharp, periodic transmission dips with adjustable free spectral range and extinction ratio. This hybrid configuration supports multifunctional behavior, enabling the device to operate as a narrowband filter, modulator, or sensor depending on the tuning parameters. The SCOW+ structure leverages self-coupling and phase interference to induce coupled-resonator-induced transparency, offering fine control over spectral features. Using frequency-domain simulations, we validate the spectral response and tunability of SCOW+. Simulation results confirm that the device exhibits flexible tuning capabilities and dynamic reconfiguration of its transmission profile by adjusting ring length and coupling coefficient. SCOW+ enhances spectral shaping without significantly increasing device size. Its modularity and compatibility with standard fabrication processes underscore its potential for scalable integration in silicon photonics platforms. The results of this study highlight the versatility of SCOW-derived architectures and enable compact, tunable photonic components in next-generation integrated systems. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

20 pages, 39376 KB  
Proceeding Paper
AI-Powered Real-Time Image Recognition System with a Laser-Based Deterrent for Primate Pest Control in Orchards
by Sung-Wen Wang, Shih-Ming Cho, Min-Chie Chiu and Shao-Chun Chen
Eng. Proc. 2026, 134(1), 65; https://doi.org/10.3390/engproc2026134065 - 21 Apr 2026
Viewed by 533
Abstract
We developed an automated system to address orchard crop damage caused by Formosan macaques, a problem where traditional deterrent methods have proven to be ineffective. The system integrates an Internet Protocol camera with a You Only Look Once version 5 (YOLOv5) object detection [...] Read more.
We developed an automated system to address orchard crop damage caused by Formosan macaques, a problem where traditional deterrent methods have proven to be ineffective. The system integrates an Internet Protocol camera with a You Only Look Once version 5 (YOLOv5) object detection model, which was trained on an augmented 6000-image dataset featuring a simulated monkey puppet in an indoor setting to validate its real-time identification capability through simulation. Upon target detection, a high-power laser, controlled via the Message Queuing Telemetry Transport protocol, is actuated to perform dynamic and non-invasive repelling. A web-based Human–Machine Interface (HMI) is provided, allowing users to remotely monitor and adjust strategies. This system offers a low-cost, highly efficient, and scalable solution for smart agriculture, with potential for expansion to other scenarios requiring a high degree of security and defense, such as warehouses and construction sites. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 11519 KB  
Proceeding Paper
Correlation Analysis Between Preparation Movements and Smash Performance in Badminton Using You Only Look Once Algorithm and Sensor Data
by Wen-Yu Lin, Wen-Huang Lin and You-Jen Lin
Eng. Proc. 2026, 134(1), 66; https://doi.org/10.3390/engproc2026134066 - 17 Apr 2026
Viewed by 174
Abstract
The badminton smash is a decisive scoring technique whose effectiveness depends on adequate preparation and a proper proximal-to-distal sequencing of the kinetic chain. This study integrates a You Only Look Once (YOLO)-based real-time vision detector with five wearable inertial measurement units (IMUs) attached [...] Read more.
The badminton smash is a decisive scoring technique whose effectiveness depends on adequate preparation and a proper proximal-to-distal sequencing of the kinetic chain. This study integrates a You Only Look Once (YOLO)-based real-time vision detector with five wearable inertial measurement units (IMUs) attached to the right shoulder, right elbow, right wrist, right hip, and right knee of right-handed players. A high-speed camera provides video for shuttlecock and joint localization via YOLO, and the IMUs provide instantaneous joint accelerations at impact. The following four coaching-oriented indicators are defined: (1) rapid lowering of the center of mass after the opponent’s shot; (2) immediate forward acceleration after the shuttle is released; (3) alignment at the hitting position with the right shoulder/hip rotated backward and the left shoulder facing the approaching shuttle; and (4) a proximal-to-distal sequence in which the shoulder leads the elbow and then the wrist. Using two athletes with 15 trials each, the system achieved an overall recognition accuracy above 93% against manually annotated video. The method can provide objective feedback for coaches and players and is suitable for instructional use in physical education classes. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1986 KB  
Proceeding Paper
Smart Cloud-Connected Near Infrared Spectroscopy Device for Non-Invasive Blood Glucose Tracking
by Joshua Mari M. Buenaventura, Jose Angelo T. Macalintal, Charmaine C. Paglinawan and Julius T. Sese
Eng. Proc. 2026, 134(1), 67; https://doi.org/10.3390/engproc2026134067 - 21 Apr 2026
Viewed by 290
Abstract
A non-invasive blood glucose monitoring system was developed in this study using near-infrared spectroscopy and an Arduino platform. Reflected signals from near-infrared-focused emissions to a user’s finger are captured via an infrared-tuned photodiode, digitally processed, and displayed on an Android-based application with logging, [...] Read more.
A non-invasive blood glucose monitoring system was developed in this study using near-infrared spectroscopy and an Arduino platform. Reflected signals from near-infrared-focused emissions to a user’s finger are captured via an infrared-tuned photodiode, digitally processed, and displayed on an Android-based application with logging, reminders, and cloud synchronization. Calibrated testing with 20 participants (10 diabetics and 10 non-diabetics) revealed that in the measurement of diabetics, the non-fasting readings showed high average accuracy (99.89%). Non-diabetic trials also demonstrated strong measurement acuity (92.18%), with improved accuracy in non-fasting measurements. The device demonstrates feasibility for affordable, portable, and cloud-connected smart non-invasive glucose tracking. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

9 pages, 1128 KB  
Proceeding Paper
Implementation of Support Vector Machine for Aroma-Based Classification of Traditional Filipino Beverages
by John Paul T. Cruz, Chris B. Domingo, Ealiezerr Andrei E. Ladia, Marites B. Tabanao and Roben A. Juanatas
Eng. Proc. 2026, 134(1), 68; https://doi.org/10.3390/engproc2026134068 - 22 Apr 2026
Viewed by 199
Abstract
This study presents an E-nose system for the identification and classification of volatile compounds in traditional Filipino alcoholic beverages, Basi, Bignay, Lambanog, and Tapuy. The system utilizes a gas sensor array composed of MQ3, MQ6, MQ8, MQ135, and MQ136 sensors, and implements a [...] Read more.
This study presents an E-nose system for the identification and classification of volatile compounds in traditional Filipino alcoholic beverages, Basi, Bignay, Lambanog, and Tapuy. The system utilizes a gas sensor array composed of MQ3, MQ6, MQ8, MQ135, and MQ136 sensors, and implements a Support Vector Machine (SVM) algorithm with principal component analysis for classification and dimensionality reduction. The experimental process involves three main phases: absorption, data acquisition, and desorption. A total of 225 training samples per class and a total of 20 testing samples were used, evenly distributed among all classes. The SVM model achieved an accuracy of 85%, highlighting its effectiveness in distinguishing between the beverages. This work contributes to the advancement of low-cost, sensor-based solutions for quality control, standardization, and the cultural preservation of traditional Filipino wines. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

14 pages, 2674 KB  
Proceeding Paper
Parameter Determination of Quantum Approximate Optimization Algorithm Using Layerwise Grid Search Method
by Su-Ling Lee and Chien-Cheng Tseng
Eng. Proc. 2026, 134(1), 69; https://doi.org/10.3390/engproc2026134069 - 22 Apr 2026
Viewed by 287
Abstract
The quantum approximate optimization algorithm (QAOA) is an efficient method for solving combinatorial optimization problems in quantum computing. These problems involve finding the best solution from a finite set of possibilities. At its core, the QAOA uses an Ansatz circuit composed of alternating [...] Read more.
The quantum approximate optimization algorithm (QAOA) is an efficient method for solving combinatorial optimization problems in quantum computing. These problems involve finding the best solution from a finite set of possibilities. At its core, the QAOA uses an Ansatz circuit composed of alternating unitary operators, the mixing and problem Hamiltonians, that are controlled by a set of parameters. Its goal is to find the optimal parameters so that the final quantum state of the circuit encodes the problem’s solution. While this parameter optimization is often handled by classical optimizers, including constrained optimization by linear approximations (COBYLA) and Nelder–Mead, these methods frequently present local extrema. Therefore, we developed a layerwise grid search (LGS) method as an alternative. Since a full grid search is too time-consuming, the LGS method significantly reduces the search time while still finding a good solution. To demonstrate its effectiveness, we present experimental results for the max-cut problem, comparing the performance of our LGS method against conventional classical optimizers. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 17245 KB  
Proceeding Paper
Image Classification of Asiatic Parakeets Using YOLOv5 and Residual Network 50
by Terenz Ace C. Flores, Francis Danielle G. Luna and Jocelyn F. Villaverde
Eng. Proc. 2026, 134(1), 70; https://doi.org/10.3390/engproc2026134070 - 22 Apr 2026
Viewed by 306
Abstract
Parakeets exhibit many similar traits across species, with only subtle differences in features and coloration used for classification, which complicates detection and identification for birdwatchers, breeders, and researchers. Traditional classification methods rely on observation, while more expensive options involve DNA sampling. We developed [...] Read more.
Parakeets exhibit many similar traits across species, with only subtle differences in features and coloration used for classification, which complicates detection and identification for birdwatchers, breeders, and researchers. Traditional classification methods rely on observation, while more expensive options involve DNA sampling. We developed a bird classification system that identifies Asiatic parakeets by combining You Only Look Once Version 5 (YOLOv5) for detection with ResNet-50 for the classification of four specific species: Alexandrine, Moustache, Plum-headed, and Indian Ringneck parakeets. Using a Raspberry Pi 4B and a Raspberry Pi Camera housed in a customized enclosure to capture images of the birds, the evaluation indicated an overall accuracy of 95.05% through a multi-class confusion matrix, demonstrating the effectiveness of the system as a reliable tool for avian identification and research. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

11 pages, 3891 KB  
Proceeding Paper
Nose Detection Based on Quadratic Curve Fitting with Geometric–Photometric–Structural Scoring
by Yu-Chen Chen, Shao-Chi Kao and Jian-Jiun Ding
Eng. Proc. 2026, 134(1), 71; https://doi.org/10.3390/engproc2026134071 - 22 Apr 2026
Viewed by 165
Abstract
An edge-based and curve-based rule-driven nose detection framework is designed to improve the reliability of face detection. The designed framework combines quadratic curve fitting with a calibrated scoring mechanism that fuses geometric, photometric, and structural information into a unified model. These stages jointly [...] Read more.
An edge-based and curve-based rule-driven nose detection framework is designed to improve the reliability of face detection. The designed framework combines quadratic curve fitting with a calibrated scoring mechanism that fuses geometric, photometric, and structural information into a unified model. These stages jointly enforce symmetry consistency, reliable tip position, and clear wing boundaries. Candidate face regions are first refined by skin filtering and ellipse validation, from which a mid-lower facial ROI is framed for nasal candidate extraction. We further incorporate eye/mouth hints (EyeMap/MouthMap) to restrict the region of interest (ROI) to the region below the eyes, above the mouth, and between the two eyes. When a mouth is detected, this ROI refinement supersedes the chrominance-red (Cr) channel trimming; otherwise, we fall back to the Cr channel horizontal projection to detect dominant mouth peaks and trim the lower-lip band, thereby suppressing lip interference. A multi-threshold Canny procedure with histogram projection is employed to collect multiple nose rectangles by selecting various vertical and horizontal peaks under three adaptive threshold scales. Within each rectangle, edge contours are quadratically fitted and categorized into U-shape (nasal base), N-shape (nostril rim), and C-shape (nasal wings), enabling rule-based selection of the base, wings, and nostrils. The fused features are then processed by a calibrated geometric–photometric–structural scoring module that uses YCbCr contrasts and red/black penalties to suppress lip and eye confounders. Experiments with diverse faces and lighting conditions show accurate and stable nose localization, with notably reliable wing fitting and nasal base detection, improving the accuracy of face detection. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 1161 KB  
Proceeding Paper
Human Event and Action Analysis Using Transformer-Based Multimodal AI
by Ralph Edcel R. Fabian, Peter Miles Anthony L. Laporre, Louis Raphael Q. Lagare, Paul Emmanuel G. Empas and John Paul T. Cruz
Eng. Proc. 2026, 134(1), 72; https://doi.org/10.3390/engproc2026134072 - 22 Apr 2026
Viewed by 226
Abstract
With the increasing demand for enhanced security and surveillance, the integration of multimodal AI has shown significant promise. We developed and fine-tuned a transformer-based model, the Large Language and Vision Assistant–OneVision, tailored for human event and action recognition. By utilizing a multimodal approach, [...] Read more.
With the increasing demand for enhanced security and surveillance, the integration of multimodal AI has shown significant promise. We developed and fine-tuned a transformer-based model, the Large Language and Vision Assistant–OneVision, tailored for human event and action recognition. By utilizing a multimodal approach, we identified specific human actions, including eating, running, fighting, sitting, and sleeping, within diverse real-world settings. Through knowledge distillation and Low-Rank Adaptation, the model’s performance was optimized in demonstrating substantial improvements in context-aware recognition and response generation. Evaluation results showed recall-oriented understudy for obtaining evaluation (ROUGE)-1 score of 0.6844, ROUGE-2 score of 0.5751, ROUGE-L score of 0.6520, and the bilingual evaluation understudy score of 68.20, demonstrating significant gains in accuracy and interpretability. The model’s success highlights its potential for real-time applications in surveillance, healthcare, and interactive AI systems, providing reliable, efficient, and context-sensitive human action detection. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 645 KB  
Proceeding Paper
Hylocereus undatus Maturity Classification Using You Only Look Once Version 7
by Adrian Q. Adajar, Nicouli Vincent V. Cagampan and Isagani V. Villamor
Eng. Proc. 2026, 134(1), 73; https://doi.org/10.3390/engproc2026134073 - 22 Apr 2026
Viewed by 206
Abstract
Dragon fruit (Hylocereus undatus) is a high-value crop in the Philippines that has gained commercial importance due to its nutritional benefits and profitability. However, determining the optimal maturity stage remains challenging for farmers relying on manual classification. We developed an automated [...] Read more.
Dragon fruit (Hylocereus undatus) is a high-value crop in the Philippines that has gained commercial importance due to its nutritional benefits and profitability. However, determining the optimal maturity stage remains challenging for farmers relying on manual classification. We developed an automated system that integrates You Only Look Once Version 7 (YOLOv7) for dragon fruit detection. A dataset of dragon fruit images across three maturity levels, unripe, ripe, and over-ripe, was collected and used to train the model. The system classifies maturity stages based on external features such as color and shape, and its performance will be evaluated using a confusion matrix. By providing accurate classification, the proposed system aims to assist farmers in harvesting dragon fruits at their optimal stage, improving yield quality and market competitiveness while reducing human error. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 1493 KB  
Proceeding Paper
Support Vector Machine-Based Electronic Nose System for Spoilage Detection in Coconut Milk-Based Filipino Foods
by John Paul T. Cruz, Pamela Nicole De Guzman, Alec Louisse Bermillo, Emmy Grace T. Requillo and Roben A. Juanatas
Eng. Proc. 2026, 134(1), 74; https://doi.org/10.3390/engproc2026134074 - 22 Apr 2026
Viewed by 323
Abstract
Coconut milk-based Filipino foods provide a favorable environment for microbial growth and are highly susceptible to spoilage. Traditionally, spoilage in such foods has been assessed through subjective sensory evaluation, a method that often lacks consistency and accuracy. The present study introduces an electronic [...] Read more.
Coconut milk-based Filipino foods provide a favorable environment for microbial growth and are highly susceptible to spoilage. Traditionally, spoilage in such foods has been assessed through subjective sensory evaluation, a method that often lacks consistency and accuracy. The present study introduces an electronic nose system employing Support Vector Machine (SVM) algorithms to objectively and quantitatively determine spoilage in coconut milk-based Filipino foods, including Bicol Express, Ginataang Langka, Laing, Bilo-bilo, Maja Blanca, and Ginumis. The developed system integrates six MQ gas sensors connected to an Arduino Nano and a Raspberry Pi 4B to detect and process volatile organic compounds emitted from the foods. The SVM algorithm was selected for its effectiveness in high-dimensional spaces and its ability to construct a binary classifier capable of distinguishing between spoiled and fresh samples. Dimensionality reduction in sensor data was achieved using Principal Component Analysis, which further enhanced classifier performance. System evaluation results demonstrated a high classification accuracy of approximately 98.95%, indicating the robustness of the proposed approach. The utilization of this technology offers significant benefits, not only for individuals with impaired olfactory function but also for the food industry, providing a reliable tool for food quality control and safety. Moreover, the outcomes suggest broader applicability to other perishable food products, with potential contributions to improved global food safety and storage practices. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 2823 KB  
Proceeding Paper
Innovative Filipino Sign Language Translation and Interpretation with MediaPipe
by Zylwyn A. Alejo, Nathan Cyvel Jann R. Fuentes, Maria Patricia Z. Lungay, Alpha Isabel D. Maniquez, Paul Emmanuel G. Empas and John Paul T. Cruz
Eng. Proc. 2026, 134(1), 75; https://doi.org/10.3390/engproc2026134075 - 22 Apr 2026
Viewed by 544
Abstract
Filipino Sign Language (FSL) serves as a vital means of communication for the Deaf and hard-of-hearing in the Philippines. However, its societal use remains limited due to the scarcity of qualified interpreters and the general lack of FSL literacy among the population. Therefore, [...] Read more.
Filipino Sign Language (FSL) serves as a vital means of communication for the Deaf and hard-of-hearing in the Philippines. However, its societal use remains limited due to the scarcity of qualified interpreters and the general lack of FSL literacy among the population. Therefore, this study aims to address the gap between FSL development and automated FSL translation by employing machine learning and computer vision techniques. A model was trained using the FSL-105 dataset, which comprises video clips of gestures related to greetings and colors, and utilized MediaPipe for real-time detection of hand, face, and body landmarks. Through iterative training with transfer learning, the model’s performance improved from an initial accuracy of 80% to a final accuracy of 98.75%. The results demonstrate that the MediaPipe-based model can reliably interpret FSL gestures, positioning it as a potentially accessible assistive tool for the Deaf and hard of hearing community. This technology holds promise for applications in education, healthcare, and public service, offering new opportunities to promote the social inclusion of Filipino Deaf communities through more inclusive communication. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

15 pages, 1302 KB  
Proceeding Paper
Quantum-Resistant Encryption for IoT Communication in Critical Engineering Infrastructure
by Wai Yie Leong
Eng. Proc. 2026, 134(1), 76; https://doi.org/10.3390/engproc2026134076 - 22 Apr 2026
Viewed by 523
Abstract
The growing interconnection of critical engineering infrastructure through IoT introduces unprecedented exposure to cyber threats. Emerging quantum computing capabilities pose a transformative risk to classical cryptographic primitives such as Rivest–Shamir–Adleman and Elliptic-Curve Cryptography, which underpin secure communication and device authentication in industrial control [...] Read more.
The growing interconnection of critical engineering infrastructure through IoT introduces unprecedented exposure to cyber threats. Emerging quantum computing capabilities pose a transformative risk to classical cryptographic primitives such as Rivest–Shamir–Adleman and Elliptic-Curve Cryptography, which underpin secure communication and device authentication in industrial control systems, power grids, transportation networks, and healthcare infrastructure. This paper investigates quantum-resistant encryption, often termed post-quantum cryptography (PQC), as a sustainable security paradigm for IoT communication within critical systems. By analyzing lattice-based, code-based, multivariate, and hash-based schemes, the study evaluates trade-offs between computational cost, memory footprint, and latency constraints intrinsic to resource-limited IoT nodes. A hybrid architectural framework integrating the National Institute of Standards and Technology-standardized algorithms (e.g., Cryptographic Suite for Algebraic Lattices—Kyber, Dilithium) with lightweight symmetric primitives (e.g., Ascon, GIFT block cipher in Combined Feedback mode) is proposed for secure data transmission across heterogeneous IoT layers. Experimental simulations benchmark key-exchange throughput, ciphertext expansion, and resilience against quantum-adversarial models, demonstrating up to 65% reduction in handshake latency compared to baseline lattice implementations under constrained conditions. The paper concludes with policy and engineering recommendations for the adoption of quantum-resistant IoT protocols in energy, transportation, and industrial automation sectors, highlighting alignment with global PQC migration roadmaps and IEC 62443 cybersecurity standards. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 897 KB  
Proceeding Paper
Implementation of Deep Belief Network with Sensor Correction Algorithm to Predict Weather on a Raspberry Pi
by Alaric S. Espiña, Franchesca Shieville F. Castro and Rosemarie V. Pellegrino
Eng. Proc. 2026, 134(1), 77; https://doi.org/10.3390/engproc2026134077 - 21 Apr 2026
Viewed by 218
Abstract
Weather is an essential part of life that affects livelihoods such as agriculture, aviation, etc. Existing systems for weather prediction use deep learning frameworks such as Recurrent Neural Networks and Long Short-term Memory. These models, however, suffer from vanishing gradients that affect the [...] Read more.
Weather is an essential part of life that affects livelihoods such as agriculture, aviation, etc. Existing systems for weather prediction use deep learning frameworks such as Recurrent Neural Networks and Long Short-term Memory. These models, however, suffer from vanishing gradients that affect the accuracy of the prediction. Using the Deep Belief Networks, we developed a model to address this. Historical weather data is obtained from the Philippine Atmospheric, Geophysical and Astronomical Services Administration for model training. The ground-level sensor data was used to normalize the inputs for the model. The resulting multiclass accuracy is 80%. A larger dataset is recommended for better performance. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

11 pages, 3080 KB  
Proceeding Paper
Aerodynamic Analysis and Design of a Hybrid-Electric Vertical Take-Off and Landing Fixed-Wing Unmanned Aerial Vehicles
by Cheng-Quan Li, Cheng-Ying Lo and Jen-Chieh Cheng
Eng. Proc. 2026, 134(1), 78; https://doi.org/10.3390/engproc2026134078 - 21 Apr 2026
Viewed by 203
Abstract
We designed a vertical take-off and landing fixed-wing unmanned aerial vehicle for long endurance. The design process involves mission-based sizing, aerodynamic shaping of major components, and performance evaluation through numerical simulation. Wing design parameters such as airfoil type, aspect ratio, taper ratio, and [...] Read more.
We designed a vertical take-off and landing fixed-wing unmanned aerial vehicle for long endurance. The design process involves mission-based sizing, aerodynamic shaping of major components, and performance evaluation through numerical simulation. Wing design parameters such as airfoil type, aspect ratio, taper ratio, and wingtip shape are analyzed for aerodynamic efficiency. The fuselage design emphasizes nose geometry to optimize internal space and reduce drag. Four tail configurations are assessed for their effects on stability and control. The final configuration adopts an MH139 airfoil, a 4° wing incidence angle, and an H-tail, fulfilling all performance and mission requirements. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 991 KB  
Proceeding Paper
Real-Time Classification of Tobacco Leaf Diseases on Raspberry Pi 5 with You Only Look Once Version 8 and Deep Convolutional Neural Network
by Miki Meliton, Joshua Carvajal and Julius Tube Sese
Eng. Proc. 2026, 134(1), 79; https://doi.org/10.3390/engproc2026134079 - 23 Apr 2026
Viewed by 248
Abstract
The accurate and timely detection of leaf diseases is essential in helping farmers take necessary corrective actions to prevent disease spread that can lead to significant crop losses, reduced yield, and economic losses. A real-time Raspberry Pi 5-based prototype classification of tobacco leaf [...] Read more.
The accurate and timely detection of leaf diseases is essential in helping farmers take necessary corrective actions to prevent disease spread that can lead to significant crop losses, reduced yield, and economic losses. A real-time Raspberry Pi 5-based prototype classification of tobacco leaf diseases using You Only Look Once Version 8 (YOLOv8n) and a Deep Convolutional Neural Network (DCNN) was developed to assist farmers with their crop disease identification. The calibration was performed by adjusting the camera mounting height and the lux level to achieve the system’s optimal performance. It was evaluated using 24 fresh tobacco leaves upon identifying the system’s optimal setting. Under optimal settings, the prototype achieved an overall accuracy of 93%, with per-class accuracies of 100% for frogeye classification, 100% for TMV classification, 94% for wildfire classification, and 78% for healthy leaves. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 952 KB  
Proceeding Paper
Obstructive Sleep Apnea (OSA) Severity Classification Using Tongue Ultrasound Images and YOLOv8
by Rosezellynda D. Regular and Cyrel O. Manlises
Eng. Proc. 2026, 134(1), 80; https://doi.org/10.3390/engproc2026134080 - 23 Apr 2026
Viewed by 292
Abstract
Obstructive sleep apnea (OSA) is a widely known sleep disorder that leads to serious health problems and complications. The standard diagnosis method of OSA is polysomnography. However, the process is time-intensive, expensive, and not readily accessible. Machine learning (ML) has been increasingly applied [...] Read more.
Obstructive sleep apnea (OSA) is a widely known sleep disorder that leads to serious health problems and complications. The standard diagnosis method of OSA is polysomnography. However, the process is time-intensive, expensive, and not readily accessible. Machine learning (ML) has been increasingly applied in various medical imaging modalities; however, there is still a lack of research on applying ML to ultrasound imaging for OSA classification. Previous studies on ML applications in medical imaging adopt X-rays, Computed Tomography, and Magnetic Resonance Imaging, leaving ultrasound as an underexplored area. Using the You-Only-Look-Once version 8 algorithm and static tongue ultrasound images, we classified OSA severity: normal, mild, moderate, and severe. A total of 280 ultrasound images were augmented to 838 images using brightness scaling, which enhanced the training process of the model. The system was tested on 60 images, achieving an overall classification accuracy of 85%. The results demonstrate the possibility and potential of using machine learning and ultrasound imaging for classifying the severity of OSA, suggesting potential assistance to clinicians in diagnosing and intervening in this condition. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

8 pages, 1931 KB  
Proceeding Paper
Maze Navigating Robot Using Lucas–Kanade Optical Flow with Coarse-to-Fine Method
by Hannah Mae Antaran and Cyrel O. Manlises
Eng. Proc. 2026, 134(1), 81; https://doi.org/10.3390/engproc2026134081 - 23 Apr 2026
Viewed by 299
Abstract
We applied the Lucas–Kanade optical flow method combined with a coarse-to-fine approach for robot navigation. While Lucas–Kanade is widely used for flow estimation and tracking, its utilization in robot navigation remains limited. Using a Raspberry Pi 5 (8 gigabytes) and a Logitech webcam, [...] Read more.
We applied the Lucas–Kanade optical flow method combined with a coarse-to-fine approach for robot navigation. While Lucas–Kanade is widely used for flow estimation and tracking, its utilization in robot navigation remains limited. Using a Raspberry Pi 5 (8 gigabytes) and a Logitech webcam, a mobile robot was developed that processes optical flow vectors to guide navigation decisions aimed at exiting a maze. While most maze navigation research relies on sensor fusion, we adopted computer vision to achieve collision-free navigation. The coarse-to-fine method effectively addresses the challenge of processing large motions inherent in Lucas–Kanade, resulting in an 80% success rate and 67% recovery rate. Simple linear regression analysis results revealed a negative correlation between optical flow magnitude and the robot’s distance to the nearest obstacle, indicating that closer obstacles correspond to higher flow magnitudes. The results highlight the potential of low-cost, vision-based autonomous navigation systems that eliminate the need for complex sensor arrays, making them suitable for cost-sensitive applications. The demonstrated effectiveness of the coarse-to-fine Lucas–Kanade method in handling large motion suggests its broader applicability in real-time robotic navigation, including autonomous vehicles and service robots operating in challenging or resource-limited environments. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1669 KB  
Proceeding Paper
Simulated Fall Detection Using a Semi-Supervised Machine Learning Method
by Julius John C. Arcilla, Ildreen D. Palaruan and Dionis A. Padilla
Eng. Proc. 2026, 134(1), 82; https://doi.org/10.3390/engproc2026134082 - 24 Apr 2026
Viewed by 212
Abstract
A multimodal strategy for fall detection within the broader domain of human activity recognition is developed in this study. A fine-tuned Inflated 3D Convolutional Network model, trained in optical flow data derived from video inputs, achieves 92.70% accuracy in classifying fall-related events. Simultaneously, [...] Read more.
A multimodal strategy for fall detection within the broader domain of human activity recognition is developed in this study. A fine-tuned Inflated 3D Convolutional Network model, trained in optical flow data derived from video inputs, achieves 92.70% accuracy in classifying fall-related events. Simultaneously, a Convolutional Neural Network–Bidirectional Long Short-Term Memory model incorporating attention mechanisms processes time-series sensor data, contributing to an ensemble performance of 97.87%. The integration of visual and sensor modalities illustrates a promising direction for developing reliable, real-time fall detection systems applicable in healthcare and assisted living environments. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 17541 KB  
Proceeding Paper
SCALEeat: Vision-Guided Food Scale for Automated Macronutrient Estimation
by Angelo Pamis Alcontin, Charls Gerald De Gala Correa and Julius Tube Sese
Eng. Proc. 2026, 134(1), 83; https://doi.org/10.3390/engproc2026134083 - 28 Apr 2026
Viewed by 288
Abstract
SCALEeat, a self-contained smart food scale, was developed to offer a convenient solution and replace manual logging with on-device recognition and weighing. The device integrated a Raspberry Pi 5, a camera, and a load cell, identifies foods and computes calories, carbohydrates, protein, and [...] Read more.
SCALEeat, a self-contained smart food scale, was developed to offer a convenient solution and replace manual logging with on-device recognition and weighing. The device integrated a Raspberry Pi 5, a camera, and a load cell, identifies foods and computes calories, carbohydrates, protein, and fat from measured weight through the Philippine Food Composition Tables (PhilFCT). Using transfer learning, a MobileNetV3-Large model trained on 25 commonly consumed items from ENNS, this achieved a 97.33% top-1 accuracy on a 300-image test set. Deployed on the prototype, SCALEeat achieved 93.60% accuracy, demonstrating practical accuracy and a lower-friction path to routine dietary assessment. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1546 KB  
Proceeding Paper
Thread Counter Using Alex Krizhevsky Convolutional Neural Network for Philippine Indigenous Textiles
by Cyris Ken M. Alipio, Paolo B. Sarmiento and Jocelyn F. Villaverde
Eng. Proc. 2026, 134(1), 84; https://doi.org/10.3390/engproc2026134084 - 27 Apr 2026
Viewed by 209
Abstract
Thread counting is used to assess the quality and cultural significance of Philippine indigenous textiles such as Kalinga and Piña. We developed a portable system that automates the process using a Raspberry Pi 4 and the Alex Krizhevsky Convolutional Neural Network. The system [...] Read more.
Thread counting is used to assess the quality and cultural significance of Philippine indigenous textiles such as Kalinga and Piña. We developed a portable system that automates the process using a Raspberry Pi 4 and the Alex Krizhevsky Convolutional Neural Network. The system processes textile images, employing AlexNet to count warp and weft threads, and displays results for real-time fabric assessment. Initial tests yielded an accuracy rate of ninety-six percent. By integrating AI and portability, this work provides a technical solution while contributing to the sustainability of cultural heritage. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 845 KB  
Proceeding Paper
You Only Look Once-Based Bitter Melon Size Classification Enhanced by Harris Corner Detection and Douglas–Peucker Algorithm
by Julian Marc B. Surara, Charles Ivan Matthew C. Nangit, Analyn N. Yumang and Charmaine C. Paglinawan
Eng. Proc. 2026, 134(1), 85; https://doi.org/10.3390/engproc2026134085 - 27 Apr 2026
Viewed by 150
Abstract
Accurate size classification remains a persistent challenge for agricultural products with irregular morphology, such as bitter melon (Momordica charantia). Proper grading is essential for fair pricing, efficient packaging, and compliance with the Association of Southeast Asian Nations and Philippine National Standards, [...] Read more.
Accurate size classification remains a persistent challenge for agricultural products with irregular morphology, such as bitter melon (Momordica charantia). Proper grading is essential for fair pricing, efficient packaging, and compliance with the Association of Southeast Asian Nations and Philippine National Standards, yet traditional manual sorting often results in inconsistencies. To address this, we introduce an automated classification framework built on the You Only Look Once Version 8 (YOLOv8) model. The system integrates Harris Corner Detection to enhance feature extraction and the Douglas–Peucker algorithm to simplify contour representations, thereby reducing noise and improving shape analysis. A dataset of Ampalaya images was trained and processed to detect and categorize fruit sizes, with evaluation conducted through a confusion matrix. Experimental results showed an overall classification accuracy of 93.75%, demonstrating that the combined approach effectively balances precision with computational efficiency. Beyond improving classification accuracy, the findings highlight the broader potential of combining deep learning and contour-based methods to advance agricultural automation, optimize post-harvest workflows, and strengthen competitiveness in both local and international markets. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1029 KB  
Proceeding Paper
Residential Smart Energy Meter with Load Forecasting Using Long Short-Term Memory and Overload Protection
by Jaimvyn Kleid D. Jardiniano, Emmanuel Freeman H. Paloma, Charmaine C. Paglinawan and Ericson D. Dimaunahan
Eng. Proc. 2026, 134(1), 86; https://doi.org/10.3390/engproc2026134086 - 24 Apr 2026
Viewed by 299
Abstract
The rapid increase in residential electricity consumption in the Philippines has underscored the need for accurate metering, predictive forecasting, and improved protection against electrical hazards. We design a residential smart energy metering system with load forecasting using Long Short-Term Memory (LSTM) and integrated [...] Read more.
The rapid increase in residential electricity consumption in the Philippines has underscored the need for accurate metering, predictive forecasting, and improved protection against electrical hazards. We design a residential smart energy metering system with load forecasting using Long Short-Term Memory (LSTM) and integrated overload protection. The system employs non-invasive SCT-013 current sensors manufactured by DFRobot and a ZMPT101B voltage sensor manufactured by Qingxian Zeming Langxi Electronic, both the SCT-013 and ZMPT101B were purchased from Circuit Rocks Philippines. The SCT-013 current sensors and ZMPT101B voltage sensors are interfaced with a Raspberry Pi to measure consumption across four residential branch circuits. Data is transmitted to a cloud-based dashboard for real-time monitoring, while a relay module automatically disconnects loads under overload conditions. To validate accuracy, the prototype was deployed for 24 h and recorded a total of 18.87 kWh compared to the 20 kWh recorded from the actual Manila Electric Company (MERALCO)-provided energy meter. The LSTM model trained on per-minute data with calendar features achieved strong predictive performance across the branches. The LSTM forecasted the load and current for the next 24 h. The forecasted current was used as the dynamic tripping value for the overload protection. The overload protection tests demonstrated reliable tripping behavior within seconds of detecting overload currents. Results confirm that the system provides accurate energy monitoring, reliable overload protection, and robust short-term load forecasting. The prototype demonstrates a cost-effective and scalable approach for enhancing residential energy management, safety, and forecasting in Philippine Households. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1892 KB  
Proceeding Paper
Analysis and Testing of Night Image Positioning System
by You-Sian Lin, Shih-Hsuan Lin, Yu-Rui Chen and Hsin-Tung Ma
Eng. Proc. 2026, 134(1), 87; https://doi.org/10.3390/engproc2026134087 - 27 Apr 2026
Viewed by 175
Abstract
We developed an image-based positioning system and evaluated its performance under nighttime conditions. The system combines GPS, inertial measurement units, and camera input to determine position. Tests were conducted under three lighting scenarios: daylight lamp, low beam, and high beam. The results show [...] Read more.
We developed an image-based positioning system and evaluated its performance under nighttime conditions. The system combines GPS, inertial measurement units, and camera input to determine position. Tests were conducted under three lighting scenarios: daylight lamp, low beam, and high beam. The results show that both daylight lamp and high-beam conditions improved positioning accuracy by up to 82%, demonstrating strong adaptability to varying lighting conditions. Additionally, the difference in correction percentage between low-beam and high-beam conditions was approximately 19.6%. The system’s robust performance suggests strong potential for integration into adaptive driving beam systems, contributing to intelligent lighting control and improved safety in autonomous driving and advanced driver-assistance applications. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 1826 KB  
Proceeding Paper
Repair of Cracked Composites and Investigation of Their Performances Under Impact Load
by Ahmet Yesil, Mete Onur Kaman, Mustafa Albayrak, Hasan Ballikaya and Mehmet Fatih Sahan
Eng. Proc. 2026, 134(1), 88; https://doi.org/10.3390/engproc2026134088 - 27 Apr 2026
Viewed by 243
Abstract
In this study, composite plates with circular holes cracked on both edges were repaired using a composite patch. Epoxy adhesive was used to bond the patch to the plate. Low-velocity impact tests were applied to repaired specimens with varying crack lengths. Compared to [...] Read more.
In this study, composite plates with circular holes cracked on both edges were repaired using a composite patch. Epoxy adhesive was used to bond the patch to the plate. Low-velocity impact tests were applied to repaired specimens with varying crack lengths. Compared to specimens without holes, the maximum reaction force of repaired specimens without cracks increased by up to 50%. When the crack length reached its maximum, the maximum impact force of repaired specimens increased by 36% compared to the specimens without holes. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

10 pages, 3713 KB  
Proceeding Paper
Development of Honey Grading System Using Computer Vision and Convolutional Neural Network
by Morris Urgel Aquino and Jocelyn Flores Villaverde
Eng. Proc. 2026, 134(1), 89; https://doi.org/10.3390/engproc2026134089 - 10 Apr 2026
Viewed by 93
Abstract
We developed a prototype for grading various classes of honey. The honey was categorized into light amber, medium amber, and dark amber. The data collected were split into training and validation using an 80:20 ratio. The pre-trained model of the Mobile Convolutional Neural [...] Read more.
We developed a prototype for grading various classes of honey. The honey was categorized into light amber, medium amber, and dark amber. The data collected were split into training and validation using an 80:20 ratio. The pre-trained model of the Mobile Convolutional Neural Network Version 2 architecture was employed. Then, the model was fine-tuned and trained, achieving the training and validation accuracy of 98.09 and 96.5%, respectively. The model was integrated into the user interface and deployed using the Raspberry Pi. The model was evaluated and yielded a 91.1% testing accuracy. Overall, the developed prototype demonstrated the feasibility of deploying it in a real-world scenario for honey grading using edge devices. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 13812 KB  
Proceeding Paper
AI Video-Based Analysis of the Volleyball Forearm Pass in Continuous Wall-Volley
by Wen Huang Lin, Wen Yu Lin and Jin Cheng Lee
Eng. Proc. 2026, 134(1), 90; https://doi.org/10.3390/engproc2026134090 (registering DOI) - 30 Apr 2026
Abstract
An AI video–based assessment system is used to analyze the volleyball forearm pass under continuous wall-volley conditions in this study. A single 120 frames per second (FPS) high-speed camera captures the athlete from a rear-oblique view. A laptop executes a You Only Look [...] Read more.
An AI video–based assessment system is used to analyze the volleyball forearm pass under continuous wall-volley conditions in this study. A single 120 frames per second (FPS) high-speed camera captures the athlete from a rear-oblique view. A laptop executes a You Only Look Once (YOLO)-based pipeline to detect the ball and human keypoints, including the shoulders, elbows, wrists, hips, knees, and ankles. From the joint angles and ball–body relative positions, three cues are quantified. The first cue is the ready posture, characterized by straight arms, downward wrist flexion, an upper arm–trunk angle of approximately 90°, and a forward-leaning center of mass. The second cue is the ball–contact point located posterior to the wrist joint. The third cue is the variation in the center of mass synchronized with the rhythm of the ball. Five athletes performed ten trials, and the predictions were compared against manual annotations, achieving greater than 95% accuracy in criterion attainment. The system outputs criterion scores and key frames to provide immediate feedback. Deployment challenges, including occlusion, viewpoint, and illumination, are discussed, along with potential extensions such as multi-camera fusion and temporal tracking. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

26 pages, 6998 KB  
Proceeding Paper
Frequency and Quality Factor Analysis of Loss Factor Addition to High-Frequency AT-Cut Quartz Resonators with Femtosecond Laser Drilling Electrode and Inverted Etching
by Zi-Gui Huang and Wei-Hsiang Lee
Eng. Proc. 2026, 134(1), 91; https://doi.org/10.3390/engproc2026134091 - 23 Apr 2026
Viewed by 61
Abstract
With the advancement of computing and transmission technologies, there has been a growing demand for quartz oscillators and resonators, whose performance is evaluated by the quality coefficient (Figure of Merit, FoM). High-frequency, miniaturized fabrication is the design goal, and process optimization and innovative [...] Read more.
With the advancement of computing and transmission technologies, there has been a growing demand for quartz oscillators and resonators, whose performance is evaluated by the quality coefficient (Figure of Merit, FoM). High-frequency, miniaturized fabrication is the design goal, and process optimization and innovative design methods need to be emphasized. Based on the 1978 Institute of Electrical and Electronics Engineers standard definition, the old design parameters of AT-cut quartz crystal sheet are retained to analyze the structural loss factor, dielectric loss factor, frequency, admittance, quality factor, and error value with the fundamental frequency increased from 76.8 to 96 MHz. In this study, COMSOL Multiphysics is used to simulate and analyze the quartz resonator by introducing the femtosecond laser quartz microvia machining technique from the literature, improving the electrode and inverted wet etching process, and incorporating the structural loss factor and dielectric loss factor into the quartz resonator model to observe the changes in the quality factor, the percentage of the quality factor error, and the values of the eigen-frequency and the error of the frequency. We analyze the trend of loss factor, frequency value, and error value and analyze the process advantages and disadvantages of femtosecond laser drilling electrodes, coated electrodes, inverted wet etching, inverted dry etching, and single-side and double-side etching to provide a reference for the design of future process components. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

6 pages, 400 KB  
Proceeding Paper
Design of Hybrid Solar Powered Composter
by Vhien Francis De Quiroz, Byron Joseph Luzame and Glenn Magwili
Eng. Proc. 2026, 134(1), 92; https://doi.org/10.3390/engproc2026134092 - 29 Apr 2026
Viewed by 109
Abstract
Food waste is a serious worldwide problem that greatly contributes to resource inefficiency and degradation of the environment. However, amidst the challenge is a phenomenal opportunity for sustainability: turning food waste into useful organic compost. Therefore, we fabricated a composting machine that converts [...] Read more.
Food waste is a serious worldwide problem that greatly contributes to resource inefficiency and degradation of the environment. However, amidst the challenge is a phenomenal opportunity for sustainability: turning food waste into useful organic compost. Therefore, we fabricated a composting machine that converts food waste into organic compost matter by using a process that decreases the moisture content to make the composting time faster. We conducted two processes: one with the juicing process and one without the juicing process. The results of the juicing process showed a decrease in the moisture content from 42% to approximately 32–34%, while when using the process without juicing, it decreased from 72 to 69%. Both processes were tested in the same time span. We conducted a t-test to determine if there was a significant difference in the means of two different sets of data, such as the moisture content of compost with and without the juicing process. The p-value of each moisture sensor was close to zero. These p-values are lower than the significance level or alpha value of 0.05, showing a significant difference between the means of the four comparison data. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

7 pages, 1632 KB  
Proceeding Paper
Laying Hens Behavior Recognition Using Computer Vision and Deep Learning
by Heidee Soliman-Cuevas and Jocelyn F. Villaverde
Eng. Proc. 2026, 134(1), 93; https://doi.org/10.3390/engproc2026134093 - 8 May 2026
Viewed by 159
Abstract
Native chicken production in the Philippines is increasing, accounting for nearly half of the total population of raised chickens. Health-conscious consumers prefer native chicken due to its lower fat content. To support this growth, the government established a breeding facility featuring 10 pens, [...] Read more.
Native chicken production in the Philippines is increasing, accounting for nearly half of the total population of raised chickens. Health-conscious consumers prefer native chicken due to its lower fat content. To support this growth, the government established a breeding facility featuring 10 pens, each housing 2 to 6 laying hens and a rooster, which began operation in November 2023. In recent months, staff observed a decline in laying performance in some pens. Because chicken behavior is a key indicator of growth and production performance, this study aims to implement a real-time laying hen activity recognition system using You Only Look Once Version 11 (YOLOv11) to classify hen behaviors into multiple categories. These include active behaviors (walking, eating, drinking, pecking, dust bathing, and preening), inactive behaviors (resting or inactivity), and environmental objects (feeders and water cans). A dataset of 464 images was collected from the breeding facility in Zamboanga City, Philippines. To capture hen behavior, a TP-Link Tapo C510W outdoor WiFi camera was mounted on the ceiling at a height of 80 cm above the ground. The model demonstrated excellent performance in detecting static objects such as feeders and water cans. Among behaviors, pecking and walking were identified as the most common, while drinking and dust bathing were relatively rare. The YOLOv11-based activity recognition system successfully achieved real-time classification of hen behaviors with strong performance across most activity classes. The system reached 95% mAP50, with particularly high accuracy in detecting static objects and distinctive behaviors, thereby providing a solid foundation for future improvements in recognizing more complex or challenging behaviors. Full article
(This article belongs to the Proceedings of The 7th Eurasia Conference on IoT, Communication and Engineering 2025 (ECICE 2025))
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-93
Previous Issue
Next Issue
Back to TopTop