Sensors

2026

Jump to: 2025, 2024, 2023, 2022, 2021

26 pages, 1224 KB

Open AccessArticle

A Generative AI–Based Technical Data Extraction Tool for IoT Application Systems

by Dezheng Kong, Nobuo Funabiki, Htoo Htoo Sandi Kyaw, I Nyoman Darma Kotama, Zihao Zhu and Alfiandi Aulia Rahmadani

Sensors 2026, 26(4), 1081; https://doi.org/10.3390/s26041081 - 7 Feb 2026

Viewed by 137

Abstract

Nowadays, Internet of Things (IoT) application systems play an essential role in smart cities, industry, healthcare, agriculture, and smart homes. For non-expert users, designing and implementing IoT application systems remains challenging, especially when configuring sensors, edge devices, and server platforms. To support configuration [...] Read more.

Nowadays, Internet of Things (IoT) application systems play an essential role in smart cities, industry, healthcare, agriculture, and smart homes. For non-expert users, designing and implementing IoT application systems remains challenging, especially when configuring sensors, edge devices, and server platforms. To support configuration tasks of IoT application systems, we have developed an AI-based setup assistance tool. However, AI models still fail to reliably support newly released or previously unseen devices, sometimes producing incomplete or erroneous outputs that may lead to configuration failures. Incorporating their technical-document information into Retrieval-Augmented Generation (RAG) is an effective way to supplement AI knowledge and improve reliability. In this paper, we propose a generative AI-based technical data extraction tool to address the challenges. It extracts essential technical information using the schema-based extraction from given PDF or HTML datasheets and converts it into a structured format suitable for AI-supported configurations. A local vector database is used to enable semantic similarity retrieval and provide document-grounded evidence for RAG-based answering, ensuring consistent support for previously unseen IoT devices. For evaluations, we applied the proposal to several sensor and device datasheets and compared extracted specifications with ground-truth values to measure accuracy and completeness. Then, we compared end-to-end configuration QA reliability against a commercial baseline (ChatPDF) using the golden benchmark. The results show that the proposed tool reliably acquires key specifications and significantly improves end-to-end configuration QA reliability. Across 960 golden QA pairs, the proposed method improves Recall from 0.636 to 0.926 and Accuracy from 0.595 to 0.807 compared with ChatPDF. Full article

► Show Figures

Figure 1

2025

Jump to: 2026, 2024, 2023, 2022, 2021

29 pages, 1677 KB

Open AccessArticle

Pairwise Diverse and Uncertain Gradient-Sampling for Similarity Retrieval

by Christoffer Löffler

Sensors 2025, 25(22), 6899; https://doi.org/10.3390/s25226899 - 12 Nov 2025

Viewed by 545

Abstract

Sports tracking produces large, unstructured trajectory datasets. The search and retrieval of interesting plays are essential parts of their analysis. Since annotations are sparse, similarity search remains the standard technique. It relies on learned lower-dimensional representations for its computational feasibility. Siamese Networks learn [...] Read more.

Sports tracking produces large, unstructured trajectory datasets. The search and retrieval of interesting plays are essential parts of their analysis. Since annotations are sparse, similarity search remains the standard technique. It relies on learned lower-dimensional representations for its computational feasibility. Siamese Networks learn dimensionality reduction from pairwise distances. However, complete training datasets are impractical to compute due to their combinatorial nature and the cost of distance calculations. Sub-sampling sacrifices representation quality for speed, leading to less meaningful search results. We propose the novel sampling technique Pairwise Diverse and Uncertain Gradient (PairDUG), which exploits the model’s gradient signals to select representative and informative pairs for training. The broad experimental study implements the method for large-scale basketball and American football datasets. The results show that PairDUG at least halves the required compute time while maintaining, or even improving, retrieval quality, and outperforms other baseline methods. Furthermore, our evaluation shows that the selected pairs’ gradient signals exhibit greater magnitude, diversity, and stability than those of any other method. This work represents a foundational contribution to pairwise distance learning. Hence, future work transfers the method not only to other sports, such as soccer, but also to complex trajectory datasets outside the sports domain. Full article

► Show Figures

Figure 1

21 pages, 6013 KB

Open AccessArticle

Very-Large-Scale Integration-Friendly Method for Vital Activity Detection with Frequency-Modulated Continuous Wave Radars

by Krzysztof Ślot, Piotr Łuczak, Paweł Kapusta, Sławomir Hausman, Arto Rantala and Jacek Flak

Sensors 2025, 25(7), 2151; https://doi.org/10.3390/s25072151 - 28 Mar 2025

Cited by 1 | Viewed by 1112

Abstract

A simple algorithm for respiratory activity detection in data produced by Frequency-Modulated Continuous-Wave (FMCW) radars is presented in this paper. The proposed computational architecture can be directly mapped onto custom digital–analog VLSI hardware, which is a unique approach in research on intelligent FMCW [...] Read more.

A simple algorithm for respiratory activity detection in data produced by Frequency-Modulated Continuous-Wave (FMCW) radars is presented in this paper. The proposed computational architecture can be directly mapped onto custom digital–analog VLSI hardware, which is a unique approach in research on intelligent FMCW sensor development, offering a potential energy-efficient data analysis solution for target applications, such as preventing human trafficking or providing life-sign detection under limited visibility. The algorithm comprises two main modules. The first one summarizes radar-produced data into a descriptor reflecting the amount of motion that occurs within appropriately determined time intervals. The second one classifies a sequence of the produced descriptors using a recurrent neural network composed of gated recurrent units. To ensure the algorithm’s implementation feasibility, an analog VLSI circuit comprising its main functional blocks has been designed, manufactured, and tested, providing constraints for neural model derivation. The adverse effects of the primary constraint, the severe restriction on admissible weight resolution, have been handled by introducing a novel training loss component and a simple mechanism for diversifying the effective weight sets of different network neurons. Experimental evaluation of the presented method, performed using the dataset of indoor recordings, indicates that the proposed simple, hardware implementation-friendly algorithm provides over 94% human detection accuracy and similar F1 scores. Full article

► Show Figures

Figure 1

2024

Jump to: 2026, 2025, 2023, 2022, 2021

14 pages, 1158 KB

Open AccessArticle

Extreme R-CNN: Few-Shot Object Detection via Sample Synthesis and Knowledge Distillation

by Shenyong Zhang, Wenmin Wang, Zhibing Wang, Honglei Li, Ruochen Li and Shixiong Zhang

Sensors 2024, 24(23), 7833; https://doi.org/10.3390/s24237833 - 7 Dec 2024

Cited by 4 | Viewed by 2333

Abstract

Traditional object detectors require extensive instance-level annotations for training. Conversely, few-shot object detectors, which are generally fine-tuned using limited data from unknown classes, tend to show biases toward base categories and are susceptible to variations within these unknown samples. To mitigate these challenges, [...] Read more.

Traditional object detectors require extensive instance-level annotations for training. Conversely, few-shot object detectors, which are generally fine-tuned using limited data from unknown classes, tend to show biases toward base categories and are susceptible to variations within these unknown samples. To mitigate these challenges, we introduce a Two-Stage Fine-Tuning Approach (TFA) named Extreme R-CNN, designed to operate effectively with extremely limited original samples through the integration of sample synthesis and knowledge distillation. Our approach involves synthesizing new training examples via instance clipping and employing various data-augmentation techniques. We enhance the Faster R-CNN architecture by decoupling the regression and classification components of the Region of Interest (RoI), allowing synthetic samples to train the classification head independently of the object-localization process. Comprehensive evaluations on the Microsoft COCO and PASCAL VOC datasets demonstrate significant improvements over baseline methods. Specifically, on the PASCAL VOC dataset, the average precision for novel categories is enhanced by up to 15 percent, while on the more complex Microsoft COCO benchmark it is enhanced by up to 6.1 percent. Remarkably, in the 1-shot scenario, the AP50 of our model exceeds that of the baseline model in the 10-shot setting within the PASCAL VOC dataset, confirming the efficacy of our proposed method. Full article

► Show Figures

Figure 1

17 pages, 872 KB

Open AccessArticle

Federated Learning in the Detection of Fake News Using Deep Learning as a Basic Method

by Kristína Machová, Marián Mach and Viliam Balara

Sensors 2024, 24(11), 3590; https://doi.org/10.3390/s24113590 - 2 Jun 2024

Cited by 4 | Viewed by 4792

Abstract

This article explores the possibilities for federated learning with a deep learning method as a basic approach to train detection models for fake news recognition. Federated learning is the key issue in this research because this kind of learning makes machine learning more [...] Read more.

This article explores the possibilities for federated learning with a deep learning method as a basic approach to train detection models for fake news recognition. Federated learning is the key issue in this research because this kind of learning makes machine learning more secure by training models on decentralized data at decentralized places, for example, at different IoT edges. The data are not transformed between decentralized places, which means that personally identifiable data are not shared. This could increase the security of data from sensors in intelligent houses and medical devices or data from various resources in online spaces. Each station edge could train a model separately on data obtained from its sensors and on data extracted from different sources. Consequently, the models trained on local data on local clients are aggregated at the central ending point. We have designed three different architectures for deep learning as a basis for use within federated learning. The detection models were based on embeddings, CNNs (convolutional neural networks), and LSTM (long short-term memory). The best results were achieved using more LSTM layers (F1 = 0.92). On the other hand, all three architectures achieved similar results. We also analyzed results obtained using federated learning and without it. As a result of the analysis, it was found that the use of federated learning, in which data were decomposed and divided into smaller local datasets, does not significantly reduce the accuracy of the models. Full article

► Show Figures

Figure 1

2023

Jump to: 2026, 2025, 2024, 2022, 2021

19 pages, 1321 KB

Open AccessArticle

SignalFormer: Hybrid Transformer for Automatic Drone Identification Based on Drone RF Signals

by Xiang Yan, Bing Han, Zhigang Su and Jingtang Hao

Sensors 2023, 23(22), 9098; https://doi.org/10.3390/s23229098 - 10 Nov 2023

Cited by 7 | Viewed by 3037

Abstract

With the growing integration of drones into various civilian applications, the demand for effective automatic drone identification (ADI) technology has become essential to monitor malicious drone flights and mitigate potential threats. While numerous convolutional neural network (CNN)-based methods have been proposed for ADI [...] Read more.

With the growing integration of drones into various civilian applications, the demand for effective automatic drone identification (ADI) technology has become essential to monitor malicious drone flights and mitigate potential threats. While numerous convolutional neural network (CNN)-based methods have been proposed for ADI tasks, the inherent local connectivity of the convolution operator in CNN models severely constrains RF signal identification performance. In this paper, we propose an innovative hybrid transformer model featuring a CNN-based tokenization method that is capable of generating T-F tokens enriched with significant local context information, and complemented by an efficient gated self-attention mechanism to capture global time/frequency correlations among these T-F tokens. Furthermore, we underscore the substantial impact of incorporating phase information into the input of the SignalFormer model. We evaluated the proposed method on two public datasets under Gaussian white noise and co-frequency signal interference conditions, The SignalFormer model achieved impressive identification accuracy of 97.57% and 98.03% for coarse-grained identification tasks, and 97.48% and 98.16% for fine-grained identification tasks. Furthermore, we introduced a class-incremental learning evaluation to demonstrate SignalFormer’s competence in handling previously unseen categories of drone signals. The above results collectively demonstrate that the proposed method is a promising solution for supporting the ADI task in reliable ways. Full article

► Show Figures

Figure 1

14 pages, 3440 KB

Open AccessArticle

Hydraulic Measuring Hoses as Pressure Signal Distortion—Mathematical Model and Results of Experimental Tests

by Klaudiusz Klarecki and Dominik Rabsztyn

Sensors 2023, 23(16), 7056; https://doi.org/10.3390/s23167056 - 9 Aug 2023

Cited by 2 | Viewed by 2493

Abstract

The article presents the results of a developed model and experimental studies of the Minimess^® hydraulic signal hose’s influence on the changes in the indications of the pressure transducer during the high dynamics of hydrostatic drives and controls. The model test results [...] Read more.

The article presents the results of a developed model and experimental studies of the Minimess^® hydraulic signal hose’s influence on the changes in the indications of the pressure transducer during the high dynamics of hydrostatic drives and controls. The model test results show that measuring hoses can be used as hardware low-pass filters during the digital recording of pressure waveforms. However, the cut-off frequency values of the measuring hoses obtained using the model are dramatically lower than those observed during the experiment. The experiment results show that the measuring hoses can only be used without any limitations to measure the average pressure value. In the case of measuring pressure waveforms, the user should carefully choose the measuring hose length. For this reason, the relationship between the measuring hose length and its cut-off frequency should be known. Full article

► Show Figures

Figure 1

12 pages, 1999 KB

Open AccessCommunication

Combination of a DC Motor Controller and Telemetry System to Optimize Energy Consumption

by Paweł Żur

Sensors 2023, 23(15), 6923; https://doi.org/10.3390/s23156923 - 3 Aug 2023

Cited by 4 | Viewed by 2106

Abstract

The paper introduces the development stages of a MOSFET-based controller for a DC brush motor. The main objective was to design a controller that could be integrated with the existing telemetry system, offering full configurability through an Android application. This controller aims to [...] Read more.

The paper introduces the development stages of a MOSFET-based controller for a DC brush motor. The main objective was to design a controller that could be integrated with the existing telemetry system, offering full configurability through an Android application. This controller aims to provide real-time analysis of data collected from the measurement system, including motor revolutions and current draw. Based on the analyzed data and the conditions set in the Android application, the controller adjusts the motor’s operational characteristics accordingly. The paper provides a comprehensive description of the controller system’s functioning. The proposed control system is particularly relevant in applications where minimizing energy consumption for driving a DC motor is of utmost importance. Full article

► Show Figures

Figure 1

16 pages, 1625 KB

Open AccessArticle

Variable Rate Point Cloud Geometry Compression Method

by Lehui Zhuang, Jin Tian, Yujin Zhang and Zhijun Fang

Sensors 2023, 23(12), 5474; https://doi.org/10.3390/s23125474 - 9 Jun 2023

Cited by 6 | Viewed by 3230

Abstract

With the development of 3D sensors technology, 3D point cloud is widely used in industrial scenes due to their high accuracy, which promotes the development of point cloud compression technology. Learned point cloud compression has attracted much attention for its excellent rate distortion [...] Read more.

With the development of 3D sensors technology, 3D point cloud is widely used in industrial scenes due to their high accuracy, which promotes the development of point cloud compression technology. Learned point cloud compression has attracted much attention for its excellent rate distortion performance. However, there is a one-to-one correspondence between the model and the compression rate in these methods. To achieve compression at different rates, a large number of models need to be trained, which increases the training time and storage space. To address this problem, a variable rate point cloud compression method is proposed, which enables the adjustment of the compression rate by the hyperparameter in a single model. To address the narrow rate range problem that occurs when the traditional rate distortion loss is jointly optimized for variable rate models, a rate expansion method based on contrastive learning is proposed to expands the bit rate range of the model. To improve the visualization effect of the reconstructed point cloud, a boundary learning method is introduced to improve the classification ability of the boundary points through boundary optimization and enhance the overall model performance. The experimental results show that the proposed method achieves variable rate compression with a large bit rate range while ensuring the model performance. The proposed method outperforms G-PCC, achieving more than 70% BD-Rate against G-PCC, and performs about, as well as the learned methods at high bit rates. Full article

► Show Figures

Figure 1

22 pages, 3435 KB

Open AccessArticle

Production Planning Using a Shared Resource Register Organized According to the Assumptions of Blockchain Technology

by Barbara Balon, Krzysztof Kalinowski and Iwona Paprocka

Sensors 2023, 23(4), 2308; https://doi.org/10.3390/s23042308 - 19 Feb 2023

Cited by 6 | Viewed by 2620

Abstract

This article presents the architecture of integration of blockchain technology (BCT) and the Internet of Things with the planning of production processes. The authors proposed a shared concept of a distributed machine database based on BCT. As part of the work, a network [...] Read more.

This article presents the architecture of integration of blockchain technology (BCT) and the Internet of Things with the planning of production processes. The authors proposed a shared concept of a distributed machine database based on BCT. As part of the work, a network of connections for the exchange of production resources was created using nodes communicating in a decentralized system, which at the same time serves as an integration of the virtual and real environment. Particular attention was focused on developing an algorithm for the efficient division of production tasks between all interested network users. BCT is used to conclude smart contracts and transactions and ensure the security of exchanged production data within shared ledgers. The proposed concept is a solution enabling a modern approach to the interdisciplinary management of production resources while maintaining the highest cybersecurity standards. Full article

► Show Figures

Figure 1

22 pages, 2209 KB

Open AccessArticle

Condition-Based Failure-Free Time Estimation of a Pump

by Grzegorz Ćwikła and Iwona Paprocka

Sensors 2023, 23(4), 1785; https://doi.org/10.3390/s23041785 - 5 Feb 2023

Cited by 5 | Viewed by 3043

Abstract

Reliable and continuous operation of the equipment is expected in the wastewater treatment plant, as any perturbations can lead to environmental pollution and the need to pay penalties. Optimization and minimization of operating costs of the pump station cannot, therefore, lead to a [...] Read more.

Reliable and continuous operation of the equipment is expected in the wastewater treatment plant, as any perturbations can lead to environmental pollution and the need to pay penalties. Optimization and minimization of operating costs of the pump station cannot, therefore, lead to a reduction in reliability but rather should be based on preventive works, the necessity of which should be foreseen. The purpose of this paper is to develop an accurate model to predict a pump’s mean time to failure, allowing for rational planning of maintenance. The pumps operate under the supervision of the automatic control system and SCADA, which is the source of historical data on pump operation parameters. This enables the research and development of various methods and algorithms for optimizing service activities. In this case, a multiple linear regression model is developed to describe the impact of historical data on pump operation for pump maintenance. In the literature, the least squares method is used to estimate unknown regression coefficients for this data. The original value of the paper is the application of the genetic algorithm to estimate coefficient values of the multiple linear regression model of failure-free time of the pump. Necessary analysis and simulations are performed on the data collected for submersible pumps in a sewage pumping station. As a result, an improvement in the adequacy of the presented model was identified. Full article

► Show Figures

Figure 1

2022

Jump to: 2026, 2025, 2024, 2023, 2021

21 pages, 515 KB

Open AccessArticle

Transient Behavior of a Queueing Model with Hyper-Exponentially Distributed Processing Times and Finite Buffer Capacity

by Wojciech M. Kempa and Iwona Paprocka

Sensors 2022, 22(24), 9909; https://doi.org/10.3390/s22249909 - 16 Dec 2022

Cited by 3 | Viewed by 1915

Abstract

In the paper, a finite-capacity queueing model is considered in which jobs arrive according to a Poisson process and are being served according to hyper-exponential service times. A system of equations for the time-sensitive queue-size distribution is established by applying the paradigm of [...] Read more.

In the paper, a finite-capacity queueing model is considered in which jobs arrive according to a Poisson process and are being served according to hyper-exponential service times. A system of equations for the time-sensitive queue-size distribution is established by applying the paradigm of embedded Markov chain and total probability law. The solution of the corresponding system written for Laplace transforms is obtained via an algebraic approach in a compact form. Numerical illustration results are attached as well. Full article

► Show Figures

Figure 1

19 pages, 3969 KB

Open AccessArticle

A Predictive Approach for Disassembly Line Balancing Problems

by Iwona Paprocka and Bożena Skołud

Sensors 2022, 22(10), 3920; https://doi.org/10.3390/s22103920 - 22 May 2022

Cited by 15 | Viewed by 3726

Abstract

In selective serial disassembly sequence planning, when the target node (component) is reached, the selective disassembly task is completed and the refurbished component is repaired, reused or remanufactured. Since the efficient utilization of existing resources is necessary, it is crucial to predict disassembly [...] Read more.

In selective serial disassembly sequence planning, when the target node (component) is reached, the selective disassembly task is completed and the refurbished component is repaired, reused or remanufactured. Since the efficient utilization of existing resources is necessary, it is crucial to predict disassembly operation times and the condition of joints for recycling, reusing or remanufacturing. The method of estimating the disassembly times of a joint if it is intended for remanufacturing, recycling and reuse is an important and urgent requirement for research development and results. The aim of the paper is to investigate the disassembly system with predicted operation times and the quality of product connections (joints) in order to balance the line smoothness index, to minimize a line time factor, line efficiency and profit and minimize an ex post error. Disassembly times for remanufacturing, recycling and reuse are estimated separately based on the historical data of disassembly times and the quality of joints. The presented estimation method of disassembly operation times increases the reliability and efficiency of elaborated balances of tasks in lines. Underestimated disassembly operation times can be compensated for during the idle points in the successive cycles, provided that the transport operations are performed manually and that travel time determines the cycle time. Full article

► Show Figures

Figure 1

20 pages, 3056 KB

Open AccessArticle

Application of Blockchain Technology in Production Scheduling and Management of Human Resources Competencies

by Barbara Balon, Krzysztof Kalinowski and Iwona Paprocka

Sensors 2022, 22(8), 2844; https://doi.org/10.3390/s22082844 - 7 Apr 2022

Cited by 26 | Viewed by 4361

Abstract

Today, enterprises are multitasking, with branches set up all over the world. Virtual enterprises are created to make better use of existing resources, improve the quality of manufactured products and agilely respond to customer requirements. In order to fully meet the requirements of [...] Read more.

Today, enterprises are multitasking, with branches set up all over the world. Virtual enterprises are created to make better use of existing resources, improve the quality of manufactured products and agilely respond to customer requirements. In order to fully meet the requirements of enterprises, a decentralized structure of data registration and transmission and authentication of network users is needed. The information collected via the Internet of Things and flowing based on the properties of the Blockchain (BC) network facilitates enterprise resource planning and enables the integration of internal processes, especially when planning, changing the current or introducing new production. The aim of this paper is to present the concept of using a common data register in BC technology, which enables a number of applications related to the automation of the process of selecting human resources for production tasks. The paper presents an analysis of the problem related to the integration of production scheduling and human resource management with blockchain technology. Also presented is a literature analysis on scheduling, blockchain technology and data storage in the blockchain network. The analysis presents how the blockchain network works and how exactly it fits into production engineering with its advantages and disadvantages. An employee evaluation method based on the resource work history and determination of its current value within individual competencies is presented. The integration of production scheduling and human resource management with the use of BC technology is simulated. The most important advantage is faster and more effective planning thanks to the elimination of all intermediary channels in the flow of production transactions. Production tasks are balanced with production capacity in entities belonging to the virtual enterprise in parallel. For future research, different online planning algorithms will be developed and compared to achieve consortium members’ consensus on production and human resources planning. Full article

► Show Figures

Figure 1

17 pages, 6534 KB

Open AccessArticle

Recreating the Motion Trajectory of a System of Articulated Rigid Bodies on the Basis of Incomplete Measurement Information and Unsupervised Learning

by Bartłomiej Nalepa, Magdalena Pawlyta, Mateusz Janiak, Agnieszka Szczęsna, Aleksander Gwiazda and Konrad Wojciechowski

Sensors 2022, 22(6), 2198; https://doi.org/10.3390/s22062198 - 11 Mar 2022

Viewed by 2757

Abstract

Re-creating the movement of an object consisting of articulated rigid bodies is an issue that concerns both mechanical and biomechanical systems. In the case of biomechanical systems, movement re-storation allows, among other things, introducing changes in training or rehabilitation exercises. Motion recording, both [...] Read more.

Re-creating the movement of an object consisting of articulated rigid bodies is an issue that concerns both mechanical and biomechanical systems. In the case of biomechanical systems, movement re-storation allows, among other things, introducing changes in training or rehabilitation exercises. Motion recording, both in the case of mechanical and biomechanical systems, can be carried out with the use of sensors recording motion parameters or vision systems and with hybrid solutions. This article presents a method of measuring motion parameters with IMU (Inertial Measurement Unit) sensors. The main assumption of the article is to present the method of data estimation from the IMU sensors for the given time moment on the basis of data from the previous time moment. The tested system was an industrial robot, because such a system allows identifying the measurement errors from IMU sensors and estimating errors basing on the reference measurements from encoders. The aim of the research is to be able to re-create the movement parameters of an object consisting of articulated rigid bodies on the basis of incomplete measurement information from sensors. The developed algorithms can be used in the diagnostics of mechanical systems as well as in sport or rehabilitation. Limiting sensors will allow, for example, athletes defining mistakes made during training only on the basis of measurements from one IMU sensor, e.g., installed in a smartphone. Both in the case of rehabilitation and sports, minimizing the number of sensors allows increasing the comfort of the person performing a given movement as part of the measurement. Full article

► Show Figures

Figure 1

2021

Jump to: 2026, 2025, 2024, 2023, 2022

26 pages, 3664 KB

Open AccessArticle

Study on Multi-Model Soft Sensor Modeling Method and Its Model Optimization for the Fermentation Process of Pichia pastoris

by Bo Wang, Xingyu Wang, Mengyi He and Xianglin Zhu

Sensors 2021, 21(22), 7635; https://doi.org/10.3390/s21227635 - 17 Nov 2021

Cited by 7 | Viewed by 3316

Abstract

The problems that the key biomass variables in Pichia pastoris fermentation process are difficult measure in real time; this paper mainly proposes a multi-model soft sensor modeling method based on the piecewise affine (PWA) modeling method, which is optimized by particle swarm optimization [...] Read more.

The problems that the key biomass variables in Pichia pastoris fermentation process are difficult measure in real time; this paper mainly proposes a multi-model soft sensor modeling method based on the piecewise affine (PWA) modeling method, which is optimized by particle swarm optimization (PSO) with an improved compression factor (ICF). Firstly, the false nearest neighbor method was used to determine the order of the PWA model. Secondly, the ICF-PSO algorithm was proposed to cooperatively optimize the number of PWA models and the parameters of each local model. Finally, a least squares support vector machine was adopted to determine the scope of action of each local model. Simulation results show that the proposed ICF-PSO-PWA multi-model soft sensor modeling method accurately approximated the nonlinear features of Pichia pastoris fermentation, and the model prediction accuracy is improved by 4.4884% compared with the weighted least squares vector regression model optimized by PSO. Full article

► Show Figures

Figure 1

19 pages, 7455 KB

Open AccessArticle

COVID-19 Case Recognition from Chest CT Images by Deep Learning, Entropy-Controlled Firefly Optimization, and Parallel Feature Fusion

by Muhammad Attique Khan, Majed Alhaisoni, Usman Tariq, Nazar Hussain, Abdul Majid, Robertas Damaševičius and Rytis Maskeliūnas

Sensors 2021, 21(21), 7286; https://doi.org/10.3390/s21217286 - 2 Nov 2021

Cited by 81 | Viewed by 5597

Abstract

In healthcare, a multitude of data is collected from medical sensors and devices, such as X-ray machines, magnetic resonance imaging, computed tomography (CT), and so on, that can be analyzed by artificial intelligence methods for early diagnosis of diseases. Recently, the outbreak of [...] Read more.

In healthcare, a multitude of data is collected from medical sensors and devices, such as X-ray machines, magnetic resonance imaging, computed tomography (CT), and so on, that can be analyzed by artificial intelligence methods for early diagnosis of diseases. Recently, the outbreak of the COVID-19 disease caused many deaths. Computer vision researchers support medical doctors by employing deep learning techniques on medical images to diagnose COVID-19 patients. Various methods were proposed for COVID-19 case classification. A new automated technique is proposed using parallel fusion and optimization of deep learning models. The proposed technique starts with a contrast enhancement using a combination of top-hat and Wiener filters. Two pre-trained deep learning models (AlexNet and VGG16) are employed and fine-tuned according to target classes (COVID-19 and healthy). Features are extracted and fused using a parallel fusion approach—parallel positive correlation. Optimal features are selected using the entropy-controlled firefly optimization method. The selected features are classified using machine learning classifiers such as multiclass support vector machine (MC-SVM). Experiments were carried out using the Radiopaedia database and achieved an accuracy of 98%. Moreover, a detailed analysis is conducted and shows the improved performance of the proposed scheme. Full article

► Show Figures

Figure 1

24 pages, 914 KB

Open AccessArticle

Application of NSGA-II to Obtain the Charging Current-Time Tradeoff Curve in Battery Based Underwater Wireless Sensor Nodes

by Daniel Rodríguez García, Juan-A. Montiel-Nelson, Tomás Bautista and Javier Sosa

Sensors 2021, 21(16), 5324; https://doi.org/10.3390/s21165324 - 6 Aug 2021

Cited by 1 | Viewed by 2357

Abstract

In this paper, a novel application of the Nondominated Sorting Genetic Algorithm II (NSGA II) is presented for obtaining the charging current–time tradeoff curve in battery based underwater wireless sensor nodes. The selection of the optimal charging current and times is a common [...] Read more.

In this paper, a novel application of the Nondominated Sorting Genetic Algorithm II (NSGA II) is presented for obtaining the charging current–time tradeoff curve in battery based underwater wireless sensor nodes. The selection of the optimal charging current and times is a common optimization problem. A high charging current ensures a fast charging time. However, it increases the maximum power consumption and also the cost and complexity of the power supply sources. This research studies the tradeoff curve between charging currents and times in detail. The design exploration methodology is based on a two nested loop search strategy. The external loop determines the optimal design solutions which fulfill the designers’ requirements using parameters like the sensor node measurement period, power consumption, and battery voltages. The inner loop executes a local search within working ranges using an evolutionary multi-objective strategy. The experiments proposed are used to obtain the charging current–time tradeoff curve and to exhibit the accuracy of the optimal design solutions. The exploration methodology presented is compared with a bisection search strategy. From the results, it can be concluded that our approach is at least four times better in terms of computational effort than a bisection search strategy. In terms of power consumption, the presented methodology reduced the required power at least 3.3 dB in worst case scenarios tested. Full article

► Show Figures

Figure 1

30 pages, 17701 KB

Open AccessArticle

Cat and Mouse Based Optimizer: A New Nature-Inspired Optimization Algorithm

by Mohammad Dehghani, Štěpán Hubálovský and Pavel Trojovský

Sensors 2021, 21(15), 5214; https://doi.org/10.3390/s21155214 - 31 Jul 2021

Cited by 91 | Viewed by 8318

Abstract

Numerous optimization problems designed in different branches of science and the real world must be solved using appropriate techniques. Population-based optimization algorithms are some of the most important and practical techniques for solving optimization problems. In this paper, a new optimization algorithm called [...] Read more.

Numerous optimization problems designed in different branches of science and the real world must be solved using appropriate techniques. Population-based optimization algorithms are some of the most important and practical techniques for solving optimization problems. In this paper, a new optimization algorithm called the Cat and Mouse-Based Optimizer (CMBO) is presented that mimics the natural behavior between cats and mice. In the proposed CMBO, the movement of cats towards mice as well as the escape of mice towards havens is simulated. Mathematical modeling and formulation of the proposed CMBO for implementation on optimization problems are presented. The performance of the CMBO is evaluated on a standard set of objective functions of three different types including unimodal, high-dimensional multimodal, and fixed-dimensional multimodal. The results of optimization of objective functions show that the proposed CMBO has a good ability to solve various optimization problems. Moreover, the optimization results obtained from the CMBO are compared with the performance of nine other well-known algorithms including Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Gravitational Search Algorithm (GSA), Teaching-Learning-Based Optimization (TLBO), Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), Marine Predators Algorithm (MPA), Tunicate Swarm Algorithm (TSA), and Teamwork Optimization Algorithm (TOA). The performance analysis of the proposed CMBO against the compared algorithms shows that CMBO is much more competitive than other algorithms by providing more suitable quasi-optimal solutions that are closer to the global optimal. Full article

► Show Figures

Figure 1

26 pages, 3448 KB

Open AccessArticle

Teamwork Optimization Algorithm: A New Optimization Approach for Function Minimization/Maximization

by Mohammad Dehghani and Pavel Trojovský

Sensors 2021, 21(13), 4567; https://doi.org/10.3390/s21134567 - 3 Jul 2021

Cited by 121 | Viewed by 5787

Abstract

Population-based optimization algorithms are one of the most widely used and popular methods in solving optimization problems. In this paper, a new population-based optimization algorithm called the Teamwork Optimization Algorithm (TOA) is presented to solve various optimization problems. The main idea in designing [...] Read more.

Population-based optimization algorithms are one of the most widely used and popular methods in solving optimization problems. In this paper, a new population-based optimization algorithm called the Teamwork Optimization Algorithm (TOA) is presented to solve various optimization problems. The main idea in designing the TOA is to simulate the teamwork behaviors of the members of a team in order to achieve their desired goal. The TOA is mathematically modeled for usability in solving optimization problems. The capability of the TOA in solving optimization problems is evaluated on a set of twenty-three standard objective functions. Additionally, the performance of the proposed TOA is compared with eight well-known optimization algorithms in providing a suitable quasi-optimal solution. The results of optimization of objective functions indicate the ability of the TOA to solve various optimization problems. Analysis and comparison of the simulation results of the optimization algorithms show that the proposed TOA is superior and far more competitive than the eight compared algorithms. Full article

► Show Figures

Figure 1

18 pages, 3371 KB

Open AccessArticle

A Compact High-Quality Image Demosaicking Neural Network for Edge-Computing Devices

by Shuyu Wang, Mingxin Zhao, Runjiang Dou, Shuangming Yu, Liyuan Liu and Nanjian Wu

Sensors 2021, 21(9), 3265; https://doi.org/10.3390/s21093265 - 8 May 2021

Cited by 16 | Viewed by 6704

Abstract

Image demosaicking has been an essential and challenging problem among the most crucial steps of image processing behind image sensors. Due to the rapid development of intelligent processors based on deep learning, several demosaicking methods based on a convolutional neural network (CNN) have [...] Read more.

Image demosaicking has been an essential and challenging problem among the most crucial steps of image processing behind image sensors. Due to the rapid development of intelligent processors based on deep learning, several demosaicking methods based on a convolutional neural network (CNN) have been proposed. However, it is difficult for their networks to run in real-time on edge computing devices with a large number of model parameters. This paper presents a compact demosaicking neural network based on the UNet++ structure. The network inserts densely connected layer blocks and adopts Gaussian smoothing layers instead of down-sampling operations before the backbone network. The densely connected blocks can extract mosaic image features efficiently by utilizing the correlation between feature maps. Furthermore, the block adopts depthwise separable convolutions to reduce the model parameters; the Gaussian smoothing layer can expand the receptive fields without down-sampling image size and discarding image information. The size constraints on the input and output images can also be relaxed, and the quality of demosaicked images is improved. Experiment results show that the proposed network can improve the running speed by 42% compared with the fastest CNN-based method and achieve comparable reconstruction quality as it on four mainstream datasets. Besides, when we carry out the inference processing on the demosaicked images on typical deep CNN networks, Mobilenet v1 and SSD, the accuracy can also achieve 85.83% (top 5) and 75.44% (mAP), which performs comparably to the existing methods. The proposed network has the highest computing efficiency and lowest parameter number through all methods, demonstrating that it is well suitable for applications on modern edge computing devices. Full article

► Show Figures

Figure 1

23 pages, 6960 KB

Open AccessEditor’s ChoiceArticle

A Modified Sparrow Search Algorithm with Application in 3d Route Planning for UAV

by Guiyun Liu, Cong Shu, Zhongwei Liang, Baihao Peng and Lefeng Cheng

Sensors 2021, 21(4), 1224; https://doi.org/10.3390/s21041224 - 9 Feb 2021

Cited by 238 | Viewed by 10599

Abstract

The unmanned aerial vehicle (UAV) route planning problem mainly centralizes on the process of calculating the best route between the departure point and target point as well as avoiding obstructions on route to avoid collisions within a given flight area. A highly efficient [...] Read more.

The unmanned aerial vehicle (UAV) route planning problem mainly centralizes on the process of calculating the best route between the departure point and target point as well as avoiding obstructions on route to avoid collisions within a given flight area. A highly efficient route planning approach is required for this complex high dimensional optimization problem. However, many algorithms are infeasible or have low efficiency, particularly in the complex three-dimensional (3d) flight environment. In this paper, a modified sparrow search algorithm named CASSA has been presented to deal with this problem. Firstly, the 3d task space model and the UAV route planning cost functions are established, and the problem of route planning is transformed into a multi-dimensional function optimization problem. Secondly, the chaotic strategy is introduced to enhance the diversity of the population of the algorithm, and an adaptive inertia weight is used to balance the convergence rate and exploration capabilities of the algorithm. Finally, the Cauchy–Gaussian mutation strategy is adopted to enhance the capability of the algorithm to get rid of stagnation. The results of simulation demonstrate that the routes generated by CASSA are preferable to the sparrow search algorithm (SSA), particle swarm optimization (PSO), artificial bee colony (ABC), and whale optimization algorithm (WOA) under the identical environment, which means that CASSA is more efficient for solving UAV route planning problem when taking all kinds of constraints into consideration. Full article

► Show Figures

Figure 1

20 pages, 4738 KB

Open AccessArticle

GENDIS: Genetic Discovery of Shapelets

by Gilles Vandewiele, Femke Ongenae and Filip De Turck

Sensors 2021, 21(4), 1059; https://doi.org/10.3390/s21041059 - 4 Feb 2021

Cited by 16 | Viewed by 5041

Abstract

In the time series classification domain, shapelets are subsequences that are discriminative of a certain class. It has been shown that classifiers are able to achieve state-of-the-art results by taking the distances from the input time series to different discriminative shapelets as the [...] Read more.

In the time series classification domain, shapelets are subsequences that are discriminative of a certain class. It has been shown that classifiers are able to achieve state-of-the-art results by taking the distances from the input time series to different discriminative shapelets as the input. Additionally, these shapelets can be visualized and thus possess an interpretable characteristic, making them appealing in critical domains, where longitudinal data are ubiquitous. In this study, a new paradigm for shapelet discovery is proposed, which is based on evolutionary computation. The advantages of the proposed approach are that: (i) it is gradient-free, which could allow escaping from local optima more easily and supports non-differentiable objectives; (ii) no brute-force search is required, making the algorithm scalable; (iii) the total amount of shapelets and the length of each of these shapelets are evolved jointly with the shapelets themselves, alleviating the need to specify this beforehand; (iv) entire sets are evaluated at once as opposed to single shapelets, which results in smaller final sets with fewer similar shapelets that result in similar predictive performances; and (v) the discovered shapelets do not need to be a subsequence of the input time series. We present the results of the experiments, which validate the enumerated advantages. Full article

► Show Figures

Figure 1

Journal Menu

Journal Browser

Artificial Intelligence in Sensors Technology

Share This Topical Collection

Editors

Topical Collection Information

Keywords

Published Papers (23 papers)

2026

Jump to: 2025, 2024, 2023, 2022, 2021

2025

Jump to: 2026, 2024, 2023, 2022, 2021

2024

Jump to: 2026, 2025, 2023, 2022, 2021

2023

Jump to: 2026, 2025, 2024, 2022, 2021

2022

Jump to: 2026, 2025, 2024, 2023, 2021

2021

Jump to: 2026, 2025, 2024, 2023, 2022

Further Information

Guidelines

MDPI Initiatives

Follow MDPI