Next Issue
Volume 14, May-2
Previous Issue
Volume 14, April-2
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.6
Journals
Electronics
Volume 14
Issue 9
share announcement

Electronics, Volume 14, Issue 9 (May-1 2025) – 210 articles

Cover Story (view full-size image): This article analyses the potential use of a line-start permanent-magnet synchronous motor in a drive system with a frequency converter that enables stable operation without internal feedback from the rotor position. The key advantage of this approach is that the motor, which typically operates in a vector control structure, can maintain stable operation even in the event of a speed sensor failure. The induced-pole PMSM contains a two-times-lower number of permanent magnets, but their volume in the motor rotor is the same due to demagnetization robustness. The article presents calculations, simulation analyses, and experimental validation under scalar control, confirming the feasibility of using this type of machine in fault-tolerant control drives. View this paper
  • 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.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
14 pages, 3436 KiB  
Article
Synchronization of Inhalation/Exhalation Ratio and Heart Rate Variability During Spontaneous Breathing
by Emi Yuda and Yutaka Yoshida
Electronics 2025, 14(9), 1903; https://doi.org/10.3390/electronics14091903 - 7 May 2025
Viewed by 202
Abstract
In this study, we investigate the relationship between breathing patterns and cardiac autonomic nervous activity during spontaneous breathing. Electrocardiograms and respiratory signals were simultaneously monitored in six subjects for 5 min while in a seated position. The inhalation/exhalation ratio (i/e) was calculated, and [...] Read more.
In this study, we investigate the relationship between breathing patterns and cardiac autonomic nervous activity during spontaneous breathing. Electrocardiograms and respiratory signals were simultaneously monitored in six subjects for 5 min while in a seated position. The inhalation/exhalation ratio (i/e) was calculated, and its variance was compared with the heart rate variability index. The results showed that inhalation time tended to be longer than exhalation time, with the inhalation-to-exhalation ratio ranging from 1.074 to 1.423. Additionally, one subject exhibited an unusually slow respiratory cycle. The inhalation/exhalation ratio was partly associated with changes in the low-frequency to high-frequency ratio (LF/HF) of heart rate variability, indicating individual differences. These findings suggest that while breathing patterns play a role in autonomic nervous system regulation and may have applications in stress and respiratory health management, there are limitations to these associations. Full article
►▼ Show Figures

Figure 1

15 pages, 1346 KiB  
Article
Gate-Level Hardware Trojan Detection Method Based on K-Hypergraph
by Jiaji He, Bingxin Lin, Qizhi Zhang and Yiqiang Zhao
Electronics 2025, 14(9), 1902; https://doi.org/10.3390/electronics14091902 - 7 May 2025
Viewed by 151
Abstract
To shorten the development cycle of integrated circuit (IC) chips, third-party IP cores (3PIPs) are widely used in the design phase; however, these 3PIPs may be untrusted, creating potential vulnerabilities. Attackers may insert hardware Trojans (HTs) into 3PIPs, resulting in the leakage of [...] Read more.
To shorten the development cycle of integrated circuit (IC) chips, third-party IP cores (3PIPs) are widely used in the design phase; however, these 3PIPs may be untrusted, creating potential vulnerabilities. Attackers may insert hardware Trojans (HTs) into 3PIPs, resulting in the leakage of critical information, alteration of circuit functions, or even physical damage to circuits. This has attracted considerable attention, leading to increased research efforts focusing on detection methods for HTs. This paper proposes a K-Hypergraph model construction methodology oriented towards the abstraction of HT characteristics, aiming at detecting HTs. This method employs the K-nearest neighbors (K-NN) algorithm to construct a hypergraph model of gate-level netlists based on the extracted features. To ensure data balance, the SMOTE algorithm is employed before constructing the K-Hypergraph model. Then, the K-Hypergraph model is trained, and the weights of the K-Hypergraph are updated to accomplish the classification task of distinguishing between Trojan nodes and normal nodes. The experimental results demonstrate that, when evaluating Trust-Hub benchmark performance indicators, the proposed method has average balanced accuracy of 91.18% in classifying Trojan nodes, with a true positive rate (TPR) of 92.12%. Full article
►▼ Show Figures

Figure 1

17 pages, 5284 KiB  
Article
Physical Layer Interface Design and Implementation for Serial Data Transmission with Multiplier Technique Approach
by Yusuf Yalçın Kardaş and Mehmet Siraç Özerdem
Electronics 2025, 14(9), 1901; https://doi.org/10.3390/electronics14091901 - 7 May 2025
Viewed by 154
Abstract
Serial communication enables communication between devices by providing high speed and efficiency in data transfer within modern communication systems. It has a wide range of applications, including business, healthcare, education, industry, and consumer electronics. This method offers an economical and efficient solution as [...] Read more.
Serial communication enables communication between devices by providing high speed and efficiency in data transfer within modern communication systems. It has a wide range of applications, including business, healthcare, education, industry, and consumer electronics. This method offers an economical and efficient solution as it requires fewer cables and consumes fewer resources compared to parallel communication. It is particularly preferred in scenarios requiring remote communication and high-speed data transmission. Although serial communication offers speed and efficiency in data transfer, it also has certain disadvantages and limitations. Various techniques have been developed over time to enhance the speed and efficiency of data transmission. Line coding techniques have also evolved within this context. In this paper, a new technique (multiplier technique) has been developed, offering a new perspective on the line coding techniques used in serial data transmission by processing repetitive data to transmit multiple data units at once. This technique, like other line coding techniques, aims to increase the data transmission rate and overcome bandwidth limitations. In multi-level line coding techniques, instead of symbol data corresponding to each level, the coded data is derived by transmitting the number of repetitions of the logic value as a multiplier. In multi-level line coding techniques, instead of symbol data corresponding to each level, the encoded data is derived by transmitting the number of repetitions of the logic value as a multiplier. In the application example, within the logic voltage level range, a microcontroller, logic gates, and analog switches were used to analyze, deduplicate, and replicate the data. In addition, an interface design and a basic-level protocol were developed for data transmission and analyzed for high data transmission efficiency based on various parameters included in the technique. Full article
(This article belongs to the Section Computer Science & Engineering)
►▼ Show Figures

Figure 1

30 pages, 13188 KiB  
Article
Research on Sensorless Control System of Permanent Magnet Synchronous Motor Based on Improved Fuzzy Super Twisted Sliding Mode Observer
by Haoran Jiang, Xiaodong Lv, Xiaoqi Fan and Guangming Zhang
Electronics 2025, 14(9), 1900; https://doi.org/10.3390/electronics14091900 - 7 May 2025
Viewed by 204
Abstract
In order to achieve precise vector control of permanent magnet synchronous motors and maintain reliability during operation, it is necessary to obtain more accurate rotor position and rotor angular velocity. However, the installation of sensors can lead to increased motor volume and cost, [...] Read more.
In order to achieve precise vector control of permanent magnet synchronous motors and maintain reliability during operation, it is necessary to obtain more accurate rotor position and rotor angular velocity. However, the installation of sensors can lead to increased motor volume and cost, so it is necessary to use sensorless estimation of rotor position and angular velocity. The switching function of traditional sliding mode observers is a discontinuous sign function, which can lead to serious chattering problems and phase lag problems caused by low-pass filters. Therefore, this article proposes an improved fuzzy hyper spiral sliding mode observer based on the traditional sliding mode observer. Firstly, the observer takes the current as the observation object and uses the difference between the actual current and the observed current and its derivative as the fuzzy input. The sliding mode gain is used as the fuzzy output to tune the parameters of the sliding mode gain. Secondly, in response to the chattering problem caused by traditional sliding mode control methods, the hyper spiral algorithm is adopted and a sin (arctan(nx)) nonlinear function is introduced instead of the sign function as the switching function to achieve switch continuous sliding mode control, thereby suppressing the system’s chattering. Finally, the rotor position information is extracted through an orthogonal normalized phase-locked loop to improve observation accuracy. For time-varying nonlinear permanent magnet synchronous motor control systems, fractional order PID can improve the control accuracy of the system and adjust the dynamic performance of the system more quickly compared to traditional PID control algorithms. Therefore, fractional order PID is used instead of traditional PID controllers. By comparing simulation experiments with traditional sliding mode observers and fuzzy improved adaptive sliding mode observers, it was proven that the improved fuzzy super spiral sliding mode observer can effectively suppress chattering and extract rotor position with higher accuracy, a faster response rate, and better dynamic performance. This provides a new approach for the sensorless control strategy of permanent magnet synchronous motors. Full article
►▼ Show Figures

Figure 1

18 pages, 2989 KiB  
Article
Interpretable Deep Learning for Pediatric Pneumonia Diagnosis Through Multi-Phase Feature Learning and Activation Patterns
by Petra Radočaj and Goran Martinović
Electronics 2025, 14(9), 1899; https://doi.org/10.3390/electronics14091899 - 7 May 2025
Viewed by 233
Abstract
Pediatric pneumonia remains a critical global health challenge requiring accurate and interpretable diagnostic solutions. Although deep learning has shown potential for pneumonia recognition on chest X-ray images, gaps persist in understanding model interpretability and feature learning during training. We evaluated four convolutional neural [...] Read more.
Pediatric pneumonia remains a critical global health challenge requiring accurate and interpretable diagnostic solutions. Although deep learning has shown potential for pneumonia recognition on chest X-ray images, gaps persist in understanding model interpretability and feature learning during training. We evaluated four convolutional neural network (CNN) architectures, i.e., InceptionV3, InceptionResNetV2, DenseNet201, and MobileNetV2, using three approaches—standard convolution, multi-scale convolution, and strided convolution—all incorporating the Mish activation function. Among the tested models, InceptionResNetV2, with strided convolutions, demonstrated the best performance, achieving an accuracy of 0.9718. InceptionV3 also performed well using the same approach, with an accuracy of 0.9684. For DenseNet201 and MobileNetV2, the multi-scale convolution approach was more effective, with accuracies of 0.9676 and 0.9437, respectively. Gradient-weighted class activation mapping (Grad-CAM) visualizations provided critical insights, e.g., multi-scale convolutions identified diffuse viral pneumonia patterns across wider lung regions, while strided convolutions precisely highlighted localized bacterial consolidations, aligning with radiologists’ diagnostic priorities. These findings establish the following architectural guidelines: strided convolutions are suited to deep hierarchical CNNs, while multi-scale approaches optimize compact models. This research significantly advances the development of interpretable, high-performance diagnostic systems for pediatric pneumonia using chest X-rays, bridging the gap between computational innovation and clinical application. Full article
(This article belongs to the Section Computer Science & Engineering)
►▼ Show Figures

Graphical abstract

20 pages, 10497 KiB  
Article
Dual Circularly Polarized Textile Antenna with Dual Bands and On-/Off-Body Communication Modes for Multifunctional Wearable Devices
by Yi Fan, Xiongying Liu, Hongcai Yang and Zhenglin Ju
Electronics 2025, 14(9), 1898; https://doi.org/10.3390/electronics14091898 - 7 May 2025
Viewed by 194
Abstract
A circularly polarized (CP) textile antenna is investigated for concurrent on- and off-body wireless communications in the 2.38 GHz medical body area network and 5.8 GHz industrial, scientific, and medical bands in the wireless body area network. The proposed scheme consists of a [...] Read more.
A circularly polarized (CP) textile antenna is investigated for concurrent on- and off-body wireless communications in the 2.38 GHz medical body area network and 5.8 GHz industrial, scientific, and medical bands in the wireless body area network. The proposed scheme consists of a square microstrip patch antenna (MPA), in which four shorting pins are employed to tune the two resonate modes of TM10 and TM00. Notably, the slant corners on MPA are cut symmetrically to realize unidirectional CP radiation, enabling off-body communication. Moreover, four rotating L-shaped parasite elements are loaded to excite the horizontal polarization mode (TMhp), which is combined with the TM00 mode to implement CP omnidirectional radiation along the human body. For verification, a proof-of-concept prototype with the dimensions of 45 mm × 45 mm × 2 mm was fabricated and characterized. The measured −10 dB impedance bandwidths of 2.5% and 6.7%, the 3 dB AR bandwidths of 2.5% and 2.7%, and the maximum realized gains of −2.8 and 6.8 dBic are achieved in dual bands, respectively. The experimental tests, such as human body loading, structural deformation, and humidity variation, were carried out. In addition, the wireless communication capability was measured and the radiation safety is evaluated. These performances show that the proposed antenna is an appropriate choice for multifunctional wearable applications. Full article
(This article belongs to the Special Issue Antenna Design and Its Applications)
►▼ Show Figures

Figure 1

22 pages, 14183 KiB  
Article
VexNet: Vector-Composed Feature-Oriented Neural Network
by Xiao Du, Ziyou Guo, Zihao Li, Yang Cao, Xing Chen and Tieru Wu
Electronics 2025, 14(9), 1897; https://doi.org/10.3390/electronics14091897 - 7 May 2025
Viewed by 138
Abstract
Extracting robust features against geometric transformations and adversarial perturbations remains a critical challenge in deep learning. Although capsule networks exhibit resilience through vector-encapsulated features and dynamic routing, they suffer from computational inefficiency due to iterative routing, dense matrix operations, and extra activation scalars. [...] Read more.
Extracting robust features against geometric transformations and adversarial perturbations remains a critical challenge in deep learning. Although capsule networks exhibit resilience through vector-encapsulated features and dynamic routing, they suffer from computational inefficiency due to iterative routing, dense matrix operations, and extra activation scalars. To address these limitations, we propose a method that integrates (1) compact vector-grouped neurons to eliminate activation scalars, (2) a non-iterative voting algorithm that preserves spatial relationships with reduced computation, and (3) efficient weight-sharing strategies that balance computational efficiency with generalizability. Our approach outperforms existing methods in image classification on CIFAR-10 and SVHN, achieving up to a 0.31% increase in accuracy with fewer parameters and lower FLOPs. Evaluations demonstrate superior performance over competing methods, with 0.31% accuracy gains on CIFAR-10/SVHN (with reduced parameters and FLOPs) and 1.93%/1.09% improvements in novel-view recognition on smallNORB. Under FGSM and BIM attacks, our method reduces attack success rates by 47.7% on CIFAR-10 and 32.4% on SVHN, confirming its enhanced robustness and efficiency. Future work will extend vexel representations to MLPs and RNNs and explore applications in computer graphics, natural language processing, and reinforcement learning. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Graphical abstract

21 pages, 2352 KiB  
Article
Weak-Cue Mixed Similarity Matrix and Boundary Expansion Clustering for Multi-Target Multi-Camera Tracking Systems in Highway Scenarios
by Sixian Chan, Shenghao Ni, Zheng Wang, Yuan Yao, Jie Hu, Xiaoxiang Chen and Suqiang Li
Electronics 2025, 14(9), 1896; https://doi.org/10.3390/electronics14091896 - 7 May 2025
Viewed by 106
Abstract
In highway scenarios, factors such as high-speed vehicle movement, lighting conditions, and positional changes significantly affect the quality of trajectories in multi-object tracking. This, in turn, impacts the trajectory clustering process within the multi-target multi-camera tracking (MTMCT) system. To address this challenge, we [...] Read more.
In highway scenarios, factors such as high-speed vehicle movement, lighting conditions, and positional changes significantly affect the quality of trajectories in multi-object tracking. This, in turn, impacts the trajectory clustering process within the multi-target multi-camera tracking (MTMCT) system. To address this challenge, we present the weak-cue mixed similarity matrix and boundary expansion clustering (WCBE) MTMCT system. First, the weak-cue mixed similarity matrix (WCMSM) enhances the original trajectory features by incorporating weak cues. Then, considering the practical scene and incorporating richer information, the boundary expansion clustering (BEC) algorithm improves trajectory clustering performance by taking the distribution of trajectory observation points into account. Finally, to validate the effectiveness of our proposed method, we conduct experiments on both the Highway Surveillance Traffic (HST) dataset developed by our team and the public CityFlow dataset. The results demonstrate promising outcomes, validating the efficacy of our approach. Full article
(This article belongs to the Special Issue Deep Learning-Based Scene Text Detection)
►▼ Show Figures

Figure 1

13 pages, 504 KiB  
Article
Construction of Hopped-Sparse Code Multiple Access Codebooks Based on Chaotic Bernoulli Frequency-Hopping Sequence
by Peiyi Zhao, Zhimin Xu and Qi Zeng
Electronics 2025, 14(9), 1895; https://doi.org/10.3390/electronics14091895 - 7 May 2025
Viewed by 104
Abstract
Traditional sparse code multiple access (SCMA) systems, which transmit user codewords through fixed subcarrier allocations, exhibit vulnerability to external jamming and interference. To address this challenge, we propose a novel SCMA codebook design incorporating the frequency-hopping (FH) technique in this paper. The construction [...] Read more.
Traditional sparse code multiple access (SCMA) systems, which transmit user codewords through fixed subcarrier allocations, exhibit vulnerability to external jamming and interference. To address this challenge, we propose a novel SCMA codebook design incorporating the frequency-hopping (FH) technique in this paper. The construction of FH-SCMA codebooks is developed by applying cyclic shifting operations to the factor graph matrix of the conventional SCMA codebooks, where the cyclic shifting patterns are governed by chaotic Bernoulli FH sequences. Through a comprehensive case study, the critical properties of the proposed FH-SCMA codebooks—the uniformity and the sparsity, along with its error-rate performance—are illustrated in detail. Through the proposed FH-SCMA codebooks, the subcarriers of FH-SCMA are randomly hopped over within the resource-block group, while retaining the sparsity requirement, thereby facilitating the multi-user detection at the receiver. The proposed FH-SCMA system (codebooks) achieves superior performance under jamming scenarios compared to both the traditional SCMA and the previous pseudo-random FH-SCMA. Full article
(This article belongs to the Section Microwave and Wireless Communications)
►▼ Show Figures

Figure 1

24 pages, 3250 KiB  
Article
Research on the Application of Single-Parent Genetic Algorithm Improved by Sine Chaotic Mapping in Parent–Child Travel Path Optimization
by Zhi-Heng Wang and Xiao-Wen Liu
Electronics 2025, 14(9), 1894; https://doi.org/10.3390/electronics14091894 - 7 May 2025
Viewed by 131
Abstract
This paper proposes a method for recommending parent–child travel destinations and planning travel routes tailored to children of different ages. The method inputs basic information about the attractions (such as ticket prices, geographical locations, opening hours, etc.) into the system database and intelligently [...] Read more.
This paper proposes a method for recommending parent–child travel destinations and planning travel routes tailored to children of different ages. The method inputs basic information about the attractions (such as ticket prices, geographical locations, opening hours, etc.) into the system database and intelligently recommends suitable attractions based on user-provided data, including the children’s age, travel time, and trip theme. The paper transforms the route planning problem into a Traveling Salesman Problem (TSP) to optimize the travel route further. It presents an improved single-parent genetic algorithm based on sine chaos mapping (SCM-SPGA) to solve and optimize the shortest path for parent–child trips. Experimental results demonstrate that this algorithm has significant advantages in path planning accuracy and efficiency. The method is applied to a tourism dataset of Hainan, providing more personalized and age-appropriate attraction recommendations for tourists planning a parent–child trip to Hainan and optimizing the travel route. The research shows that the proposed method can effectively meet the personalized needs of parent–child travelers, significantly improving the overall travel experience by offering more tailored, efficient, and enjoyable trip-planning solutions. Full article
(This article belongs to the Special Issue Advanced Machine Learning, Pattern Recognition, and Deep Learning Technologies: Methodologies and Applications)
►▼ Show Figures

Figure 1

17 pages, 127269 KiB  
Article
A Novel 28-GHz Meta-Window for Millimeter-Wave Indoor Coverage
by Chun Yang, Chuanchuan Yang, Cheng Zhang and Hongbin Li
Electronics 2025, 14(9), 1893; https://doi.org/10.3390/electronics14091893 - 7 May 2025
Viewed by 110
Abstract
Millimeter-wave signals experience substantial path loss when penetrating common building materials, hindering seamless indoor coverage from outdoor networks. To address this limitation, we present the 28-GHz “Meta-Window”, a mass-producible, visible transparent device designed to enhance millimeter-wave signal focusing. Fabricated via metal sputtering and [...] Read more.
Millimeter-wave signals experience substantial path loss when penetrating common building materials, hindering seamless indoor coverage from outdoor networks. To address this limitation, we present the 28-GHz “Meta-Window”, a mass-producible, visible transparent device designed to enhance millimeter-wave signal focusing. Fabricated via metal sputtering and etching on a standard soda-lime glass substrate, the meta-window incorporates subwavelength metallic structures arranged in a rotating pattern based on the Pancharatnam–Berry phase principle, enabling 0–360° phase control within the 25–32 GHz frequency band. A 210 mm × 210 mm prototype operating at 28 GHz was constructed using a 69 × 69 array of metasurface unit cells, leveraging planar electromagnetic lens principles. Experimental results demonstrate that the meta-window achieves greater than 20 dB signal focusing gain between 26 and 30 GHz, consistent with full-wave electromagnetic simulations, while maintaining up to 74.93% visible transmittance. This dual transparency—for both visible light and millimeter-wave frequencies—was further validated by a communication prototype system exhibiting a greater than 20 dB signal-to-noise ratio improvement and successful demodulation of a 64-QAM single-carrier signal (1 GHz bandwidth, 28 GHz) with an error vector magnitude of 4.11%. Moreover, cascading the meta-window with a reconfigurable reflecting metasurface antenna array facilitates large-angle beam steering; stable demodulation (error vector magnitude within 6.32%) was achieved within a ±40° range using the same signal parameters. Compared to conventional transmissive metasurfaces, this approach leverages established glass manufacturing techniques and offers potential for direct building integration, providing a promising solution for improving millimeter-wave indoor penetration and coverage. Full article
►▼ Show Figures

Figure 1

17 pages, 4831 KiB  
Article
Achieving Low-Latency, High-Throughput Online Partial Particle Identification for the NA62 Experiment Using FPGAs and Machine Learning
by Pierpaolo Perticaroli, Roberto Ammendola, Andrea Biagioni, Carlotta Chiarini, Andrea Ciardiello, Paolo Cretaro, Ottorino Frezza, Francesca Lo Cicero, Michele Martinelli, Roberto Piandani, Luca Pontisso, Mauro Raggi, Cristian Rossi, Francesco Simula, Matteo Turisini, Piero Vicini and Alessandro Lonardo
Electronics 2025, 14(9), 1892; https://doi.org/10.3390/electronics14091892 - 7 May 2025
Viewed by 102
Abstract
FPGA-RICH is an FPGA-based online partial particle identification system for the NA62 experiment employing AI techniques. Integrated between the readout of the Ring Imaging Cherenkov detector (RICH) and the low-level trigger processor (L0TP+), FPGA-RICH implements a fast pipeline to process in real-time the [...] Read more.
FPGA-RICH is an FPGA-based online partial particle identification system for the NA62 experiment employing AI techniques. Integrated between the readout of the Ring Imaging Cherenkov detector (RICH) and the low-level trigger processor (L0TP+), FPGA-RICH implements a fast pipeline to process in real-time the RICH raw hit data stream, producing trigger primitives containing elaborate physics information—e.g., the number of charged particles in a physics event—that L0TP+ can use to improve trigger decision efficiency. Deployed on a single FPGA, the system combines classical online processing with a compact Neural Network algorithm to achieve efficient event classification while managing the challenging ∼10 MHz throughput requirement of NA62. The streaming pipeline ensures ∼1 μs latency, comparable to that of the NA62 detectors, allowing its seamless integration in the existing TDAQ setup as an additional detector. Development leverages High-Level Synthesis (HLS) and the open-source hls4ml package software–hardware codesign workflow, enabling fast and flexible reprogramming, debugging, and performance optimization. We describe the implementation of the full processing pipeline, the Neural Network classifier, their functional validation, performance metrics and the system’s current status and outlook. Full article
(This article belongs to the Special Issue Emerging Applications of FPGAs and Reconfigurable Computing System)
►▼ Show Figures

Figure 1

21 pages, 1189 KiB  
Article
Energy-Efficient Federated Learning-Driven Intelligent Traffic Monitoring: Bayesian Prediction and Incentive Mechanism Design
by Ye Wang, Mengqi Sui, Tianle Xia, Miao Liu, Jie Yang and Haitao Zhao
Electronics 2025, 14(9), 1891; https://doi.org/10.3390/electronics14091891 - 7 May 2025
Viewed by 93
Abstract
With the growing integration of the Internet of Things (IoT), low-altitude intelligent networks, and vehicular networks, smart city traffic systems are gradually evolving into an air–ground integrated intelligent monitoring framework. However, traditional centralized model training faces challenges such as high network load due [...] Read more.
With the growing integration of the Internet of Things (IoT), low-altitude intelligent networks, and vehicular networks, smart city traffic systems are gradually evolving into an air–ground integrated intelligent monitoring framework. However, traditional centralized model training faces challenges such as high network load due to massive data transmission, energy management difficulties for mobile devices like UAVs, and privacy risks associated with non-anonymized road operation data. Therefore, this paper proposes an air–ground collaborative federated learning framework that integrates Bayesian prediction and an incentive mechanism to achieve privacy protection and communication optimization through localized model training and differentiated incentive strategies. Simulation experiments demonstrate that, compared to the Equal Contribution Algorithm (ECA) and the Importance Contribution Algorithm (ICA), the proposed method improves model convergence speed while reducing incentive costs, providing theoretical support for the reliable operation of large-scale intelligent traffic monitoring systems. Full article
►▼ Show Figures

Figure 1

18 pages, 1193 KiB  
Article
GFANet: An Efficient and Accurate Water Segmentation Network
by Shiyu Xie and Lishan Jia
Electronics 2025, 14(9), 1890; https://doi.org/10.3390/electronics14091890 - 7 May 2025
Viewed by 88
Abstract
Accurate water body detection is essential for autonomous navigation and operational planning of unmanned surface vehicles (USVs). To address model adaptability to ambiguous boundaries caused by diverse scenarios and climatic conditions, this study proposes GFANet (Global–Local Feature Attention Network) for the real-time water [...] Read more.
Accurate water body detection is essential for autonomous navigation and operational planning of unmanned surface vehicles (USVs). To address model adaptability to ambiguous boundaries caused by diverse scenarios and climatic conditions, this study proposes GFANet (Global–Local Feature Attention Network) for the real-time water surface semantic segmentation of camera-captured images. First, a Global–Local Feature (GLF) extraction module is proposed, integrating a self-attention-based local feature extractor and a multi-scale global feature extractor for parallel feature learning, thereby enhancing hierarchical feature representation. Second, a Gated Attention (GA) module is designed with a dual-branch gating mechanism to implement noise suppression and efficient low-level feature utilization. The method was validated on three publicly available datasets in relevant domains. The experimental results on the Riwa dataset show that GFANet achieves state-of-the-art segmentation performance (4.41 M parameters, 7.15 GFLOPs) with an mIoU of 82.29% and an mPA of 89.49%. Comparable performance metrics were obtained on the USVInland and WaterSeg datasets. Additionally, GFANet achieves a 154.98 FPS processing speed, meeting real-time segmentation requirements. The experimental results verify that GFANet achieves an optimal balance between high segmentation accuracy and real-time processing efficiency. Full article
►▼ Show Figures

Figure 1

19 pages, 4129 KiB  
Article
Study on an Improved YOLOv7-Based Algorithm for Human Head Detection
by Dong Wu, Weidong Yan and Jingli Wang
Electronics 2025, 14(9), 1889; https://doi.org/10.3390/electronics14091889 - 7 May 2025
Viewed by 139
Abstract
In response to the decreased accuracy in person detection caused by densely populated areas and mutual occlusions in public spaces, a human head-detection approach is employed to assist in detecting individuals. To address key issues in dense scenes—such as poor feature extraction, rough [...] Read more.
In response to the decreased accuracy in person detection caused by densely populated areas and mutual occlusions in public spaces, a human head-detection approach is employed to assist in detecting individuals. To address key issues in dense scenes—such as poor feature extraction, rough label assignment, and inefficient pooling—we improved the YOLOv7 network in three aspects: adding attention mechanisms, enhancing the receptive field, and applying multi-scale feature fusion. First, a large amount of surveillance video data from crowded public spaces was collected to compile a head-detection dataset. Then, based on YOLOv7, the network was optimized as follows: (1) a CBAM attention module was added to the neck section; (2) a Gaussian receptive field-based label-assignment strategy was implemented at the junction between the original feature-fusion module and the detection head; (3) the SPPFCSPC module was used to replace the multi-space pyramid pooling. By seamlessly uniting CBAM, RFLAGauss, and SPPFCSPC, we establish a novel collaborative optimization framework. Finally, experimental comparisons revealed that the improved model’s accuracy increased from 92.4% to 94.4%; recall improved from 90.5% to 93.9%; and inference speed increased from 87.2 frames per second to 94.2 frames per second. Compared with single-stage object-detection models such as YOLOv7 and YOLOv8, the model demonstrated superior accuracy and inference speed. Its inference speed also significantly outperforms that of Faster R-CNN, Mask R-CNN, DINOv2, and RT-DETRv2, markedly enhancing both small-object (head) detection performance and efficiency. Full article
►▼ Show Figures

Figure 1

15 pages, 8219 KiB  
Article
A Hierarchical Voltage Control Strategy for Distribution Networks Using Distributed Energy Storage
by Chao Ma, Wenjie Xiong, Zhiyuan Tang, Ziwei Li, Yonghua Xiong and Qibo Wang
Electronics 2025, 14(9), 1888; https://doi.org/10.3390/electronics14091888 - 6 May 2025
Viewed by 211
Abstract
This paper presents a novel hierarchical voltage control framework for distribution networks to mitigate voltage violations by coordinating distributed energy storage systems (DESSs). The framework establishes a two-layer architecture that integrates centralized optimization with distributed execution. In the upper layer, a model predictive [...] Read more.
This paper presents a novel hierarchical voltage control framework for distribution networks to mitigate voltage violations by coordinating distributed energy storage systems (DESSs). The framework establishes a two-layer architecture that integrates centralized optimization with distributed execution. In the upper layer, a model predictive control (MPC)-based controller computes optimal power dispatch trajectories for critical buses, effectively decoupling slow-timescale optimization from real-time adjustments. In the lower layer, a broadcast-based controller dispatches parameterized power regulation signals, enabling autonomous active power tracking by the DESS units. This hierarchical design explicitly addresses the scalability limitations of conventional centralized control and the cyber vulnerabilities of peer-to-peer distributed strategies. The effectiveness of the proposed control framework is verified on the modified IEEE 34-bus and 123-bus test feeder. The results show that the proposed method can mitigate the average voltage violation by 93.7% and show control robustness even under 60% communication loss condition. Full article
(This article belongs to the Special Issue New Advances in Distributed Control Systems and Their Latest Applications)
►▼ Show Figures

Figure 1

19 pages, 29370 KiB  
Article
Enhancing Intelligent Robot Perception with a Zero-Shot Detection Framework for Corner Casting
by Elven Kee, Jun Jie Chong, Zi Jie Choong and Michael Lau
Electronics 2025, 14(9), 1887; https://doi.org/10.3390/electronics14091887 - 6 May 2025
Viewed by 448
Abstract
This study presents a zero-shot object detection framework for corner casting detection in shipping container operations, leveraging edge computing for intelligent robotic perception and control. The proposed system integrates Grounding DINO on a Raspberry Pi, utilizing Referring Expression Comprehension (REC) and Additional Feature [...] Read more.
This study presents a zero-shot object detection framework for corner casting detection in shipping container operations, leveraging edge computing for intelligent robotic perception and control. The proposed system integrates Grounding DINO on a Raspberry Pi, utilizing Referring Expression Comprehension (REC) and Additional Feature Keywords (AFKs) to enable precise corner casting localization without model retraining. This approach reduces computational overhead while ensuring real-time deployment suitability for robotics applications. A comparative evaluation against three SSD-based models—SSD320 MobileNet-V2 FPNLite, MobileNet-V2, and EfficientDet-Lite0—reveals that Grounding DINO achieves a 7.14% higher detection score. Furthermore, a statistical effect size analysis using Cohen’s d (d = 2.2) confirms a significant performance advantage, reinforcing Grounding DINO’s efficacy in zero-shot scenarios. These findings underscore the potential of LLM-driven object detection in resource-constrained environments, offering a scalable and adaptable solution for intelligent perception and control in robotics. Full article
(This article belongs to the Special Issue Intelligent Perception and Control for Robotics)
►▼ Show Figures

Figure 1

16 pages, 1756 KiB  
Article
Multi-Scale Parallel Enhancement Module with Cross-Hierarchy Interaction for Video Emotion Recognition
by Lianqi Zhang, Yuan Sun, Jiansheng Guan, Shaobo Kang, Jiangyin Huang and Xungao Zhong
Electronics 2025, 14(9), 1886; https://doi.org/10.3390/electronics14091886 - 6 May 2025
Viewed by 147
Abstract
Video emotion recognition faces significant challenges due to the strong spatiotemporal coupling of dynamic expressions and the substantial variations in cross-scale motion patterns (e.g., subtle facial micro-expressions versus large-scale body gestures). Traditional methods, constrained by limited receptive fields, often fail to effectively balance [...] Read more.
Video emotion recognition faces significant challenges due to the strong spatiotemporal coupling of dynamic expressions and the substantial variations in cross-scale motion patterns (e.g., subtle facial micro-expressions versus large-scale body gestures). Traditional methods, constrained by limited receptive fields, often fail to effectively balance multi-scale correlations between local cues (e.g., transient facial muscle movements) and global semantic patterns (e.g., full-body gestures). To address this, we propose an enhanced attention module integrating multi-dilated convolution and dynamic feature weighting, aimed at improving spatiotemporal emotion feature extraction. Building upon conventional attention mechanisms, the module introduces a multi-branch parallel architecture. Convolutional kernels with varying dilation rates (1, 3, 5) are designed to hierarchically capture cross-scale the spatiotemporal features of low-scale facial micro-motion units (e.g., brief lip tightening), mid-scale composite expression patterns (e.g., furrowed brows combined with cheek raising), and high-scale limb motion trajectories (e.g., sustained arm-crossing). A dynamic feature adapter is further incorporated to enable context-aware adaptive fusion of multi-source heterogeneous features. We conducted extensive ablation studies and experiments on popular benchmark datasets such as the VideoEmotion-8 and Ekman-6 datasets. Experiments demonstrate that the proposed method enhances joint modeling of low-scale cues (e.g., fragmented facial muscle dynamics) and high-scale semantic patterns (e.g., emotion-coherent body language), achieving stronger cross-database generalization. Full article
►▼ Show Figures

Figure 1

27 pages, 15024 KiB  
Article
Tools for Researching the Parameters of Photovoltaic Modules
by Milan Belik, Oleksandr Rubanenko, Iryna Hunko, Olena Rubanenko, Serhii Baraban and Andriy Semenov
Electronics 2025, 14(9), 1885; https://doi.org/10.3390/electronics14091885 - 6 May 2025
Viewed by 141
Abstract
This paper addresses critical challenges in renewable energy research, particularly under the difficult operational conditions caused by the military conflict in Ukraine. Despite significant infrastructure loss due to the armed conflict (13% of solar and 70% of wind power), Ukraine maintains a commitment [...] Read more.
This paper addresses critical challenges in renewable energy research, particularly under the difficult operational conditions caused by the military conflict in Ukraine. Despite significant infrastructure loss due to the armed conflict (13% of solar and 70% of wind power), Ukraine maintains a commitment to reach 27% renewable energy in final consumption by 2030. However, the wartime conditions present unique challenges to scientific research, with laboratories vulnerable to missile strikes and frequently requiring evacuation. This paper introduces innovative portable laboratory stands designed for comprehensive analysis and monitoring of photovoltaic (PV) module parameters. These portable platforms, integrating advanced microcontrollers, sensors, and data-processing units, enable effective real-time monitoring and parameter estimation of PV modules, thereby enhancing their operational efficiency and reliability. Two distinct portable laboratory setups were developed and are detailed: the first focuses on real-time voltage and current measurements, while the second, termed the photovoltaic module parameter scanner (SPFEM), emphasizes data collection, remote data transmission, and database integration for subsequent analysis. This research provides essential tools for ensuring continuity in scientific activities and practical training for students and researchers amidst the ongoing security threats. The presented systems significantly contribute to optimizing the performance of PV systems in Ukraine and underscore the necessity for continuous adaptation and technological advancement in renewable energy infrastructure. Full article
(This article belongs to the Special Issue Next-Generation Instrumentation and IoT-Enhanced Monitoring for Advanced Power Electronics Systems)
►▼ Show Figures

Figure 1

12 pages, 482 KiB  
Article
Interactive Heritage: The Role of Artificial Intelligence in Digital Museums
by Matina Kiourexidou and Sofia Stamou
Electronics 2025, 14(9), 1884; https://doi.org/10.3390/electronics14091884 - 6 May 2025
Viewed by 468
Abstract
Museum use of artificial intelligence (AI) is becoming increasingly common, but its contribution to museum attendance is yet to be confirmed. This paper investigates whether the adoption of AI impacts museum visitation using data from 19 museums. Statistical analyses, including ANOVA and Spearman [...] Read more.
Museum use of artificial intelligence (AI) is becoming increasingly common, but its contribution to museum attendance is yet to be confirmed. This paper investigates whether the adoption of AI impacts museum visitation using data from 19 museums. Statistical analyses, including ANOVA and Spearman correlation, were conducted to determine if the use of AI has significant effects on visitors. The findings indicate no statistically significant difference between museums that use AI and those that do not (ANOVA: p = 0.263, F = 1.34), but the Spearman correlation (r = 0.448, p = 0.055) indicates a moderate positive correlation that is not statistically significant. The findings suggest that AI enhances visitor experience rather than increasing attendance. Additionally, this study proposes a conceptual framework for AI prototyping in museums. The study contributes to the ongoing debate on AI in cultural institutions by emphasizing that future research should incorporate longitudinal studies and qualitative visitor feedback in order to capture the overall impact of AI on engagement and sustainability in museums. Full article
(This article belongs to the Special Issue Advances in HCI Research)
►▼ Show Figures

Figure 1

19 pages, 2532 KiB  
Article
Achieving High Efficiency in Schnorr-Based Multi-Signature Applications in Blockchain
by Peng Zhang, Fa Ge, Zujie Tang and Weixin Xie
Electronics 2025, 14(9), 1883; https://doi.org/10.3390/electronics14091883 - 6 May 2025
Viewed by 149
Abstract
Multi-signature applications allow multiple signers to collaboratively generate a single signature on the same message, which is widely applied in blockchain to reduce the percentage of signatures in blocks and improve the throughput of transactions. The k-sum attacks are one of the [...] Read more.
Multi-signature applications allow multiple signers to collaboratively generate a single signature on the same message, which is widely applied in blockchain to reduce the percentage of signatures in blocks and improve the throughput of transactions. The k-sum attacks are one of the major challenges in designing secure multi-signature schemes. In this work, we address k-sum attacks from a novel angle by defining a Public Third Party (PTP), which is an automatic process that can be verifiable by the public and restricts the signing phase from continuing until receiving commitments from all signers. Further, a two-round multi-signature scheme HEMS with PTP is proposed, which is secure based on the discrete logarithm assumption in the random oracle model. As each signer communicates directly with the PTP instead of other co-signers, the total amount of communication is significantly reduced. In addition, as PTP participates in the computation of the aggregation and signing algorithms, the computation cost left for each signer and verifier remains the same as the basis Schnorr signature. To the best of our knowledge, this is the high efficiency that a Schnorr-based multi-signature scheme can achieve. Further, HEMS is applied in a blockchain platform, e.g., Fabric, to improve transaction efficiency. Full article
(This article belongs to the Special Issue Recent Advances in Cybersecurity and Information Security)
►▼ Show Figures

Figure 1

23 pages, 42153 KiB  
Article
Automatic Pruning and Quality Assurance of Object Detection Datasets for Autonomous Driving
by Kana Kim, Vijay Kakani and Hakil Kim
Electronics 2025, 14(9), 1882; https://doi.org/10.3390/electronics14091882 - 6 May 2025
Viewed by 225
Abstract
Large amounts of high-quality data are required to train artificial intelligence (AI) models; however, curating such data through human intervention remains cumbersome, time-consuming, and error-prone. In particular, erroneous annotations and statistical imbalances in object detection datasets can significantly degrade model performance in real-world [...] Read more.
Large amounts of high-quality data are required to train artificial intelligence (AI) models; however, curating such data through human intervention remains cumbersome, time-consuming, and error-prone. In particular, erroneous annotations and statistical imbalances in object detection datasets can significantly degrade model performance in real-world autonomous driving scenarios. This study proposes an automated pruning framework and quality assurance strategy for 2D object detection datasets to address these issues. The framework is composed of two stages: (1) noisy label identification and deletion based on labeling scores derived from the inference results of multiple object detection models, and (2) statistical distribution whitening based on class and bounding box size diversity metrics. The proposed method was designed in accordance with the ISO/IEC 25012 data quality standards to ensure data consistency, accuracy, and completeness. Experiments were conducted on widely used autonomous driving datasets, including KITTI, Waymo, nuScenes, and large-scale publicly available datasets from South Korea. An automated data pruning process was employed to eliminate anomalous and redundant samples, resulting in a more reliable and compact dataset for model training. The results demonstrate that the proposed method substantially reduces the amount of training data required, while enhancing the detection performance and minimizing manual inspection efforts. Full article
(This article belongs to the Special Issue Development and Advances in Autonomous Driving Technology)
►▼ Show Figures

Figure 1

19 pages, 1063 KiB  
Article
Enhancing Out-of-Distribution Detection Under Covariate Shifts: A Full-Spectrum Contrastive Denoising Framework
by Dengye Pan, Bin Sheng and Xiaoqiang Li
Electronics 2025, 14(9), 1881; https://doi.org/10.3390/electronics14091881 - 6 May 2025
Viewed by 213
Abstract
Out-of-distribution (OOD) detection is crucial for identifying samples that deviate from the training distribution, thereby enhancing the reliability of deep neural network models. However, existing OOD detection methods primarily address semantic shifts, where an image’s inherent semantics have changed, and often overlook covariate [...] Read more.
Out-of-distribution (OOD) detection is crucial for identifying samples that deviate from the training distribution, thereby enhancing the reliability of deep neural network models. However, existing OOD detection methods primarily address semantic shifts, where an image’s inherent semantics have changed, and often overlook covariate shifts, which are prevalent in real-world scenarios. For instance, variations in image contrast, lighting, or viewpoints can alter input features while keeping the semantic content intact. To address this, we propose the Full-Spectrum Contrastive Denoising (FSCD) framework, which improves OOD detection under covariate shifts. FSCD first establishes a robust semantic boundary and then refines feature representations through fine-tuning. Specifically, FSCD employs a dual-level perturbation augmentation module to simulate covariate shifts and a feature contrastive denoising module to effectively distinguish in-distribution samples from OOD samples. Extensive experiments on three benchmarks demonstrate that FSCD achieves state-of-the-art performance, with AUROC improvements of up to 0.51% on DIGITS, 0.55% on OBJECTS, and 2.09% on COVID compared to the previous best method while also maintaining the highest classification accuracy on covariate-shifted in-distribution samples. Full article
►▼ Show Figures

Figure 1

25 pages, 5388 KiB  
Article
Design of a Universal Safety Control Computer for Aerostats
by Yong Hao, Zhaojie Li, Yanchu Yang, Qianqian Du and Baocheng Wang
Electronics 2025, 14(9), 1880; https://doi.org/10.3390/electronics14091880 - 6 May 2025
Viewed by 177
Abstract
Amid rapid global aviation development and increasingly stringent safety standards, aerostats demonstrate vast potential in environmental monitoring, communication relay, cargo transportation, and other applications. However, their operational safety has become a critical focus. These systems face complex flight environments and dynamic mission requirements [...] Read more.
Amid rapid global aviation development and increasingly stringent safety standards, aerostats demonstrate vast potential in environmental monitoring, communication relay, cargo transportation, and other applications. However, their operational safety has become a critical focus. These systems face complex flight environments and dynamic mission requirements that demand exceptionally high safety control standards. As the core component, the safety control computer directly determines the overall safety and stability of aerostat operations. This study employed a systems engineering methodology integrating hardware selection, software architecture design, fault diagnosis, and fault tolerance to develop a universal safety control computer system with high reliability, robust real-time performance, and adaptive capabilities. By adopting high-performance processors, redundant design techniques, and modular software programming, the system significantly enhanced anti-interference performance and fault recovery capabilities. These improvements ensured precise and rapid safety control monitoring under diverse operational conditions. Experimental validation demonstrated the system’s effectiveness in supporting both remote and autonomous safety control modes, substantially mitigating flight risks. This technological breakthrough provides robust technical support for the large-scale development and safe operation of universal aerostat systems, while offering valuable insights for safety control system design in other aerospace vehicles. Full article
►▼ Show Figures

Figure 1

19 pages, 2924 KiB  
Article
An Efficient Multiple Empirical Kernel Learning Algorithm with Data Distribution Estimation
by Jinbo Huang , Zhongmei Luo  and Xiaoming Wang 
Electronics 2025, 14(9), 1879; https://doi.org/10.3390/electronics14091879 - 5 May 2025
Viewed by 257
Abstract
The Multiple Random Empirical Kernel Learning Machine (MREKLM) typically generates multiple empirical feature spaces by selecting a limited group of samples, which helps reduce training duration. However, MREKLM does not incorporate data distribution information during the projection process, leading to inconsistent performance and [...] Read more.
The Multiple Random Empirical Kernel Learning Machine (MREKLM) typically generates multiple empirical feature spaces by selecting a limited group of samples, which helps reduce training duration. However, MREKLM does not incorporate data distribution information during the projection process, leading to inconsistent performance and issues with reproducibility. To address this limitation, we introduce a within-class scatter matrix that leverages the distribution of samples, resulting in the development of the Fast Multiple Empirical Kernel Learning Incorporating Data Distribution Information (FMEKL-DDI). This approach enables the algorithm to incorporate sample distribution data during projection, improving the decision boundary and enhancing classification accuracy. To further minimize sample selection time, we employ a border point selection technique utilizing locality-sensitive hashing (BPLSH), which helps in efficiently picking samples for feature space development. The experimental results from various datasets demonstrate that FMEKL-DDI significantly improves classification accuracy while reducing training duration, thereby providing a more efficient approach with strong generalization performance. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Figure 1

19 pages, 17474 KiB  
Article
Transforming Pediatric Healthcare with Generative AI: A Hybrid CNN Approach for Pneumonia Detection
by Sotir Sotirov, Daniela Orozova, Boris Angelov, Evdokia Sotirova and Magdalena Vylcheva
Electronics 2025, 14(9), 1878; https://doi.org/10.3390/electronics14091878 - 5 May 2025
Viewed by 306
Abstract
Pneumonia is one of the leading causes of morbidity and mortality in children, making its early detection critical for effective treatment. The objective of this study is to develop and evaluate a hybrid deep learning framework that combines convolutional neural networks with intuitionistic [...] Read more.
Pneumonia is one of the leading causes of morbidity and mortality in children, making its early detection critical for effective treatment. The objective of this study is to develop and evaluate a hybrid deep learning framework that combines convolutional neural networks with intuitionistic fuzzy estimators to enhance the accuracy, sensitivity, and robustness of pneumonia detection in pediatric chest X-rays. The main background is the use of intuitionistic fuzzy estimators (IFEs). The hybrid model integrates the powerful feature extraction capabilities of CNNs with the uncertainty handling and decision-making strengths of intuitionistic fuzzy logic. By incorporating an IFE, the model is better equipped to deal with ambiguity and noise in medical imaging data, resulting in more accurate and robust pneumonia detection. Experimental results on pediatric chest X-ray datasets demonstrate the effectiveness of the proposed method, achieving higher sensitivity and specificity compared to traditional CNN approaches. The hybrid system achieved a classification accuracy of 94.93%, confirming its strong diagnostic performance. In conclusion, this hybrid model offers a promising tool to assist healthcare professionals in the early and accurate diagnosis of pneumonia in children. Full article
(This article belongs to the Special Issue Transforming Healthcare with Generative AI)
►▼ Show Figures

Figure 1

18 pages, 814 KiB  
Article
Multi-Scale Edge-Guided Image Forgery Detection via Improved Self-Supervision and Self-Adversarial Training
by Huacong Zhang, Jishen Zeng and Jianquan Yang
Electronics 2025, 14(9), 1877; https://doi.org/10.3390/electronics14091877 - 5 May 2025
Viewed by 274
Abstract
Image forgery detection, as an essential technique for analyzing image credibility, has experienced significant advancements recently. However, the forgery detection performance remains unsatisfactory in terms of meeting practical requirements. This is partly attributed to the limited availability of pixel-level annotated forgery samples and [...] Read more.
Image forgery detection, as an essential technique for analyzing image credibility, has experienced significant advancements recently. However, the forgery detection performance remains unsatisfactory in terms of meeting practical requirements. This is partly attributed to the limited availability of pixel-level annotated forgery samples and insufficient utilization of forgery traces. We try to mitigate these issues through three aspects: training data, network design, and training strategy. In the aspect of training data, we introduce iterative self-supervision which helps generate a large collection of pixel-level labeled single or composite forgery samples through one or more rounds of random copy-move, splicing, and inpainting, addressing the insufficient availability of forgery samples. In the aspect of network design, recognizing that characteristic anomalies are generally apparent at the boundary between true and fake regions, often aligning with image edges, we propose a new edge-guided learning module to effectively capture forgery traces at image edges. In the aspect of training strategy, we introduce progressive self-adversarial training, dynamically generating adversarial samples by gradually increasing the frequency and intensity of adversarial actions during training. This increases the detection difficulty, driving the detector to identify forgery traces from harder samples while maintaining a low computational cost. Comprehensive experiments have shown that the proposed method surpasses the leading competing methods, improving image-level forgery identification by 6.6% (from 73.8% to 80.4% on average F1 score) and pixel-level forgery localization by 15.2% (from 59.1% to 74.3% in average F1 score). Full article
►▼ Show Figures

Figure 1

18 pages, 2425 KiB  
Article
Toward Real-Time Posture Classification: Reality Check
by Hongbo Zhang, Denis Gračanin, Wenjing Zhou, Drew Dudash and Gregory Rushton
Electronics 2025, 14(9), 1876; https://doi.org/10.3390/electronics14091876 - 5 May 2025
Viewed by 278
Abstract
Fall prevention has always been a crucial topic for injury prevention. Research shows that real-time posture monitoring and subsequent fall prevention are important for the prevention of fall-related injuries. In this research, we determine a real-time posture classifier by comparing classical and deep [...] Read more.
Fall prevention has always been a crucial topic for injury prevention. Research shows that real-time posture monitoring and subsequent fall prevention are important for the prevention of fall-related injuries. In this research, we determine a real-time posture classifier by comparing classical and deep machine learning classifiers in terms of their accuracy and robustness for posture classification. For this, multiple classical classifiers, including classical machine learning, support vector machine, random forest, neural network, and Adaboost methods, were used. Deep learning methods, including LSTM and transformer, were used for posture classification. In the experiment, joint data were obtained using an RGBD camera. The results show that classical machine learning posture classifier accuracy was between 75% and 99%, demonstrating that the use of classical machine learning classification alone is sufficient for real-time posture classification even with missing joints or added noise. The deep learning method LSTM was also effective in classifying the postures with high accuracy, despite incurring a significant computational overhead cost, thus compromising the real-time posture classification performance. The research thus shows that classical machine learning methods are worthy of our attention, at least, to consider for reuse or reinvention, especially for real-time posture classification tasks. The insight of using a classical posture classifier for large-scale human posture classification is also given through this research. Full article
(This article belongs to the Special Issue Real-Time Computer Vision)
►▼ Show Figures

Figure 1

21 pages, 1981 KiB  
Article
Enhanced Financial Fraud Detection Using an Adaptive Voted Perceptron Model with Optimized Learning and Error Reduction
by Muhammad Binsawad
Electronics 2025, 14(9), 1875; https://doi.org/10.3390/electronics14091875 - 5 May 2025
Viewed by 312
Abstract
Financial fraud detection is an important field in financial technology, and strong and effective machine learning (ML) models are needed to detect fraudulent transactions with high accuracy and reliability. Conventional fraud detection models, like probabilistic, instance-based, and tree-based models, tend to have high [...] Read more.
Financial fraud detection is an important field in financial technology, and strong and effective machine learning (ML) models are needed to detect fraudulent transactions with high accuracy and reliability. Conventional fraud detection models, like probabilistic, instance-based, and tree-based models, tend to have high error rates, class imbalance problems, and poor adaptability to changing fraud patterns. These issues call for sophisticated methods that improve predictive accuracy while being computationally efficient. To overcome these limitations, this research introduces the Voted Perceptron (VP) model, which utilizes an iterative learning process to dynamically adapt decision boundaries based on misclassified examples. In contrast to traditional models with static decision rules, the VP model constantly updates its weight parameters, thus providing better fraud detection abilities. The evaluation compares VP with state-of-the-art machine learning models, such as Average One Dependency Estimator (A1DE), K-nearest Neighbor (KNN), Naïve Bayes (NB), Random Tree (RT), and Functional Tree (FT), by using important performance metrics, like Mean Absolute Error (MAE), Root Mean Square Error (RMSE), True Positive Rate (TPR), recall, and accuracy. Experimental results show that VP outperforms its rivals significantly, yielding better fraud detection performance with low error rates and high recall. Furthermore, an ablation study confirms the influence of essential VP model elements on general classification performance. These results demonstrate VP to be an extremely effective model for detecting financial fraud, with enhanced flexibility towards evolving fraud patterns, and confirm the necessity for intelligent fraud detection mechanisms within financial organizations. Full article
►▼ Show Figures

Figure 1

28 pages, 13553 KiB  
Article
Implementing High-Speed Object Detection and Steering Angle Prediction for Self-Driving Control
by Bao Rong Chang, Hsiu-Fen Tsai and Jia-Sian Syu
Electronics 2025, 14(9), 1874; https://doi.org/10.3390/electronics14091874 - 4 May 2025
Viewed by 302
Abstract
In the previous work, we proposed LWGSE-YOLOv4-tiny and LWDSG-ResNet18, leveraging depthwise separable and Ghost Convolutions for fast self-driving control while achieving a detection speed of 24.9 FPS. However, the system fell short of Level 4 autonomous driving safety requirements. That is, the control [...] Read more.
In the previous work, we proposed LWGSE-YOLOv4-tiny and LWDSG-ResNet18, leveraging depthwise separable and Ghost Convolutions for fast self-driving control while achieving a detection speed of 24.9 FPS. However, the system fell short of Level 4 autonomous driving safety requirements. That is, the control response speed of object detection integrated with steering angle prediction must exceed 39.2 FPS. This study enhances YOLOv11n with dual convolution and RepGhost bottleneck, forming DuCRG-YOLOv11n, significantly improving the object detection speed while maintaining accuracy. Similarly, DuC-ResNet18 improves steering angle prediction speed and accuracy. Our approach achieves 50.7 FPS, meeting Level 4 safety standards. Compared to previous work, DuCRG-YOLOv11n boosts feature extraction speed by 912.97%, while DuC-ResNet18 enhances prediction speed by 45.37% and accuracy by 12.26%. Full article
(This article belongs to the Special Issue Object Detection in Autonomous Driving)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-210
Previous Issue
Next Issue
Back to TopTop