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) – 70 articles

  • 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:
30 pages, 6350 KiB  
Article
Modular Multilevel Converter Control Strategy for AC Fault Current Maximization and Grid Code Compliance
by Ricardo Vidal-Albalate, Enrique Belenguer and Francisco Magraner
Electronics 2025, 14(9), 1763; https://doi.org/10.3390/electronics14091763 - 25 Apr 2025
Abstract
This paper proposes a dynamic current limit for modular multilevel converters (MMCs) that maximizes the injection of current during grid faults in order to mitigate the voltage dip, reduce voltage imbalances in case of an asymmetrical fault, and ensure the proper operation of [...] Read more.
This paper proposes a dynamic current limit for modular multilevel converters (MMCs) that maximizes the injection of current during grid faults in order to mitigate the voltage dip, reduce voltage imbalances in case of an asymmetrical fault, and ensure the proper operation of protective relays. The reduced short-circuit capacity of MMCs, and power converters in general, is one of their main limitations. In the event of a fault, the converter’s current is significantly lower than that of the synchronous generators, which may impact both the performance of power system protective relays and the mitigation of voltage drops during faults. Usually, to protect the MMCs themselves, their output current is limited by their control. However, the current flowing through the power semiconductors is the arm current, not the output current, and this consists of an AC and a DC component. A new current saturation strategy aiming at maximizing fault current injection, in compliance with the most recent grid codes, is proposed. This strategy limits the arm currents by dynamically adjusting the output current limit while injecting reactive currents (both positive- and negative-sequence) and active current according to the grid codes, the fault type, and voltage sag level. A theoretical analysis is carried out to determine the maximum current injection that will not exceed the arm limits, and this is then validated through detailed PSCAD simulations. With the proposed strategy, the supplied current can be increased by approximately 40%. Full article
(This article belongs to the Special Issue Control and Optimization Technologies in Renewable Energy and Integrated Energy Systems)
18 pages, 1083 KiB  
Article
A Large Language Model-Based Approach for Data Lineage Parsing
by Zhangti Li, Wenbin Guo, Yabing Gao, Di Yang and Lin Kang
Electronics 2025, 14(9), 1762; https://doi.org/10.3390/electronics14091762 - 25 Apr 2025
Abstract
The core driver of enterprise operations is data, making data lineage crucial for data management. It not only tracks data flow but also links data sources, workflows, applications, and decision-making, improving efficiency and governance. However, current data lineage parsing methods face challenges like [...] Read more.
The core driver of enterprise operations is data, making data lineage crucial for data management. It not only tracks data flow but also links data sources, workflows, applications, and decision-making, improving efficiency and governance. However, current data lineage parsing methods face challenges like high costs, long development cycles, and poor generalization, especially for non-SQL scripts. In this paper, we introduce an innovative approach leveraging pre-trained large language models (LLMs) to overcome these bottlenecks in data lineage parsing. LLMs are employed across the entire parsing pipeline, encompassing prompt construction, lineage extraction, and result standardization. Specifically, this study developed a few-shot prompting method incorporating error cases to optimize parsing performance across various types of scripts. Additionally, a collaborative Chain of Thought (CoT) and multi-expert prompting framework was designed to further enhance parsing accuracy at the operator level. The proposed approach was empirically validated using LLMs of different parameter scales on datasets comprising multiple script types (SQL, Python, Shell, Flume, etc.). The experimental results show that LLMs with 10 billion and 100 billion parameters achieved over 95% accuracy in table-level lineage parsing when utilizing the newly designed prompts. Furthermore, 100-billion-parameter LLMs exhibited substantial accuracy improvements at the operator level. Our method reinforces the feasibility and practicality in advancing data lineage parsing methodologies. Full article
(This article belongs to the Special Issue Advances in Data-Driven Artificial Intelligence)
17 pages, 2268 KiB  
Article
Modeling and Recognition of Latent False Data Injection Attacks on Distributed Cluster Control of Distribution Network
by Jinxin Ouyang, Yujie Chen, Yanbo Diao and Fei Huang
Electronics 2025, 14(9), 1761; https://doi.org/10.3390/electronics14091761 - 25 Apr 2025
Abstract
The cyber–physical deep integration in distribution networks containing distributed generators (DGs) enables false data injection attacks (FDIAs) through data tampering in monitoring systems, posing cross-domain threats to power system security. Effective FDIA identification methods remain unavailable for distribution networks containing DGs under distributed [...] Read more.
The cyber–physical deep integration in distribution networks containing distributed generators (DGs) enables false data injection attacks (FDIAs) through data tampering in monitoring systems, posing cross-domain threats to power system security. Effective FDIA identification methods remain unavailable for distribution networks containing DGs under distributed cluster control. The cyber–physical interaction characteristics are systematically analyzed in the cyber–physical system (CPS) of distribution networks containing DGs under distributed cluster control. A latent FDIA model is established with a specific attack penetration pattern being revealed. The operational impact mechanism of latent FDIAs on distributed cluster control is theoretically elucidated. A synchronous detection signal-based localization methodology is developed for latent FDIAs, coupled with an attack signal computation algorithm. These innovations enable dynamic attack identification and proactive signal isolation, thereby facilitating subsequent defense operations by system operators. Simulation results confirm the effectiveness of the proposed dynamic identification method for latent FDIAs in distribution networks containing DGs. Full article
(This article belongs to the Special Issue Latest Advances in Distributed Systems and Networked Control)
►▼ Show Figures

Figure 1

22 pages, 1299 KiB  
Article
Ground-Moving Target Relocation for a Lightweight Unmanned Aerial Vehicle-Borne Radar System Based on Doppler Beam Sharpening Image Registration
by Wencheng Liu, Zhen Chen, Zhiyu Jiang, Yanlei Li, Yunlong Liu, Xiangxi Bu and Xingdong Liang
Electronics 2025, 14(9), 1760; https://doi.org/10.3390/electronics14091760 - 25 Apr 2025
Abstract
With the rapid development of lightweight unmanned aerial vehicles (UAVs), the combination of UAVs and ground-moving target indication (GMTI) radar systems has received great interest. However, because of size, weight, and power (SWaP) limitations, the UAV may not be able to equip a [...] Read more.
With the rapid development of lightweight unmanned aerial vehicles (UAVs), the combination of UAVs and ground-moving target indication (GMTI) radar systems has received great interest. However, because of size, weight, and power (SWaP) limitations, the UAV may not be able to equip a highly accurate inertial navigation system (INS), which leads to reduced accuracy in the moving target relocation. To solve this issue, we propose using an image registration algorithm, which matches a Doppler beam sharpening (DBS) image of detected moving targets to a synthetic aperture radar (SAR) image containing coordinate information. However, when using conventional SAR image registration algorithms such as the SAR scale-invariant feature transform (SIFT) algorithm, additional difficulties arise. To overcome these difficulties, we developed a new image-matching algorithm, which first estimates the errors of the UAV platform to compensate for geometric distortions in the DBS image. In addition, to showcase the relocation improvement achieved with the new algorithm, we compared it with the affine transformation and second-order polynomial algorithms. The findings of simulated and real-world experiments demonstrate that our proposed image transformation method offers better moving target relocation results under low-accuracy INS conditions. Full article
(This article belongs to the Special Issue New Challenges in Remote Sensing Image Processing)
16 pages, 5442 KiB  
Article
Secure Retrieval of Brain Tumor Images Using Perceptual Encryption in Cloud-Assisted Scenario
by Ijaz Ahmad, Md Shahriar Uzzal and Seokjoo Shin
Electronics 2025, 14(9), 1759; https://doi.org/10.3390/electronics14091759 - 25 Apr 2025
Abstract
Scarcity of data is one of the major challenges in developing automatic computer-aided diagnosis systems, training radiologists and supporting medical research. One solution toward this is community cloud storage, which can be utilized by organizations with a common interest as a shared data [...] Read more.
Scarcity of data is one of the major challenges in developing automatic computer-aided diagnosis systems, training radiologists and supporting medical research. One solution toward this is community cloud storage, which can be utilized by organizations with a common interest as a shared data repository for joint projects and collaboration. In this large database, relevant images are often searched by an image retrieval system, for which the computation and storage capabilities of a cloud server can bring the benefits of high scalability and availability. However, the main limitation in availing third party-provided services comes from the associated privacy concerns during data transmission, storage and computation. To ensure privacy, this study implements a content-based image retrieval application for finding different types of brain tumors in the encrypted domain. In this framework, we propose a perceptual encryption technique to protect images in such a way that the features necessary for high-dimensional representation can still be extracted from the cipher images. Also, it allows data protection on the client side; therefore, the server stores and receives images in an encrypted form and has no access to the secret key information. Experimental results show that compared with conventional secure techniques, our proposed system reduced the difference in non-secure and secure retrieval performance by up to 3%. Full article
(This article belongs to the Special Issue Security and Privacy in Networks)
28 pages, 832 KiB  
Article
BranchCloak: Mitigating Side-Channel Attacks on Directional Branch Predictors
by Jihoon Kim, Hyerean Jang and Youngjoo Shin
Electronics 2025, 14(9), 1758; https://doi.org/10.3390/electronics14091758 - 25 Apr 2025
Abstract
The emerging threat of side-channel attacks targeting branch predictors on recent Intel processors has become a growing concern. These attacks rely on exploiting a pattern history table (PHT) as a source of side-channel information. Since the PHT is shared among logical cores, attackers [...] Read more.
The emerging threat of side-channel attacks targeting branch predictors on recent Intel processors has become a growing concern. These attacks rely on exploiting a pattern history table (PHT) as a source of side-channel information. Since the PHT is shared among logical cores, attackers can observe a state in the PHT entry that collides with the victim, enabling them to leak the control flow information of a victim process. Any state changes caused by the victim will reveal whether the victim’s target branch has been taken or not. In this paper, we present BranchCloak, a novel software-based mitigation technique for PHT-based side-channel attacks. The main idea of BranchCloak is to obfuscate the PHT state by augmenting the victim’s program with some r-branches near the target branch. The r-branch is a conditional branch instruction that has the following properties: (1) it collides with the target branch in the PHT, and (2) its branching decision is made uniformly at random. BranchCloak can successfully mitigate the attack without hardware modification of the vulnerable processors. By performing extensive experiments with practical applications, we show that the performance overhead of BranchCloak is negligible. Full article
(This article belongs to the Section Computer Science & Engineering)
►▼ Show Figures

Figure 1

21 pages, 11631 KiB  
Article
Torque Ripple Reduction in Switched Reluctance Machines Considering Phase Torque-Generation Capability
by Shijie Chai, Xiaoqiang Guo, Zhiyu Liu, Peng Zhang, Yueheng Ding and Wei Hua
Electronics 2025, 14(9), 1757; https://doi.org/10.3390/electronics14091757 - 25 Apr 2025
Abstract
In this paper, an improved online torque compensation strategy considering phase torque-generation capability is proposed to enhance the conventional torque-sharing function (TSF), thus reducing torque ripple for switched reluctance machines (SRMs). The improvements are mainly attributed to two aspects: First, the phase turn-on [...] Read more.
In this paper, an improved online torque compensation strategy considering phase torque-generation capability is proposed to enhance the conventional torque-sharing function (TSF), thus reducing torque ripple for switched reluctance machines (SRMs). The improvements are mainly attributed to two aspects: First, the phase turn-on angle and TSF starting angle are separated. Thus, the phase turn-on angle can be advanced independently to enhance the torque-generation capability of the incoming phase. Second, to generate the desired torque with minimum current, the torque per ampere (TPA) characteristic is considered for commutation region separation. This can ensure that in each separated region, the phase with a stronger torque-tracking ability is utilized for torque error compensation. Accordingly, efficiency is not sacrificed. In addition to improving the TSF, a direct instantaneous torque control (DITC) method combined with a PWM regulator is proposed to reduce large torque increments due to the limited control frequency. As a result, the torque ripple can be further suppressed. Finally, an experimental setup is established, and tests are conducted under different working conditions. The results demonstrate the effectiveness of the proposed method. Full article
►▼ Show Figures

Figure 1

13 pages, 7970 KiB  
Article
Investigation of a 220 GHz Traveling-Wave Tube Based upon a Flat-Roofed Sine Waveguide with a Coupling Structure
by Shuanzhu Fang, Ruixiang Xie, Jun Luo, Zhizhe Wang, Tieyang Wang and Fangfang Song
Electronics 2025, 14(9), 1756; https://doi.org/10.3390/electronics14091756 - 25 Apr 2025
Abstract
This paper presents the design and investigation of a two-stage flat-roofed sine waveguide (SWG) traveling-wave tube (TWT) incorporating a novel coupling structure. Initially, the slow-wave structure (SWS) of a 220 GHz flat-roofed SWG was optimized, and the output performance of the corresponding TWT [...] Read more.
This paper presents the design and investigation of a two-stage flat-roofed sine waveguide (SWG) traveling-wave tube (TWT) incorporating a novel coupling structure. Initially, the slow-wave structure (SWS) of a 220 GHz flat-roofed SWG was optimized, and the output performance of the corresponding TWT was thoroughly analyzed. Subsequently, a specialized coupling structure was designed and fabricated, with the experimental results demonstrating an excellent agreement with the simulation predictions. The coupling structure exhibits low reflection and is easily manufacturable, making it highly suitable for energy coupling in two-stage TWTs. Finally, a two-stage TWT, integrating both the optimized flat-roofed SWG structure and the coupling structure, was developed and characterized. Under operating conditions of a 20.8 kV beam voltage, 150 mA current, and 150 mW input power, the proposed TWT achieved remarkable performance metrics: a maximum output power of 160 W within the frequency range 210–230 GHz and a 3 dB bandwidth exceeding 20 GHz. This research provides a valuable reference solution for the realization of high-power, broadband terahertz radiation sources, contributing significantly to the advancement of terahertz vacuum electronic devices. Full article
(This article belongs to the Special Issue Vacuum Electronics: From Micro to Nano)
►▼ Show Figures

Figure 1

20 pages, 5233 KiB  
Article
Improvement of Self-Consumption Rates by Cogeneration and PV Production for Renewable Energy Communities
by Samuele Branchetti, Carlo Petrovich, Nicola Gessa and Gianluca D’Agosta
Electronics 2025, 14(9), 1755; https://doi.org/10.3390/electronics14091755 - 25 Apr 2025
Abstract
The goal of decarbonization has driven the adoption of several intervention strategies across Europe, including the promotion of Renewable Energy Communities (RECs). This study analyses an electric REC in Italy to explore the performance of different potential energy mixes combining a biogas-based cogeneration [...] Read more.
The goal of decarbonization has driven the adoption of several intervention strategies across Europe, including the promotion of Renewable Energy Communities (RECs). This study analyses an electric REC in Italy to explore the performance of different potential energy mixes combining a biogas-based cogeneration (CHP) system and photovoltaic (PV) plants. The analysis is based on a real REC composed of 53 members (mainly companies) with a Self-Sufficiency Rate (SSR) of 92% and a Self-Consumption Rate (SCR) of 60%. Adding 550 residential consumers (apartments) to the REC, the total production matches total consumption, and both SSR and SCR converge to 84%. Compared to RECs that rely solely on PV systems, this case study shows that biogas integration leads to an increase of around 40 percentage points in both SSR and SCR—equivalent to an average gain of 0.4 to 0.6 percentage points for each percentage point increase in the CHP share of the CHP-PV production mix. The analysis quantifies how SSR and SCR vary not only with different biogas/PV production ratios but, more importantly, with variations in the total annual production-to-consumption ratio of the RECs. These results can guide the design of RECs tailored to the specific characteristics of local contexts. Full article
(This article belongs to the Special Issue Smart Energy Communities: State of the Art and Future Developments)
►▼ Show Figures

Figure 1

32 pages, 6398 KiB  
Article
Big Data-Driven Distributed Machine Learning for Scalable Credit Card Fraud Detection Using PySpark, XGBoost, and CatBoost
by Leonidas Theodorakopoulos, Alexandra Theodoropoulou, Anastasios Tsimakis and Constantinos Halkiopoulos
Electronics 2025, 14(9), 1754; https://doi.org/10.3390/electronics14091754 - 25 Apr 2025
Abstract
This study presents an optimization for a distributed machine learning framework to achieve credit card fraud detection scalability. Due to the growth in fraudulent activities, this research implements the PySpark-based processing of large-scale transaction datasets, integrating advanced machine learning models: Logistic Regression, Decision [...] Read more.
This study presents an optimization for a distributed machine learning framework to achieve credit card fraud detection scalability. Due to the growth in fraudulent activities, this research implements the PySpark-based processing of large-scale transaction datasets, integrating advanced machine learning models: Logistic Regression, Decision Trees, Random Forests, XGBoost, and CatBoost. These have been evaluated in terms of scalability, accuracy, and handling imbalanced datasets. Key findings: Among the most promising models for complex and imbalanced data, XGBoost and CatBoost promise close-to-ideal accuracy rates in fraudulent transaction detection. PySpark will be instrumental in scaling these systems to enable them to perform distributed processing, real-time analysis, and adaptive learning. This study further discusses challenges like overfitting, data access, and real-time implementation with potential solutions such as ensemble methods, intelligent sampling, and graph-based approaches. Future directions are underlined by deploying these frameworks in live transaction environments, leveraging continuous learning mechanisms, and integrating advanced anomaly detection techniques to handle evolving fraud patterns. The present research demonstrates the importance of distributed machine learning frameworks for developing robust, scalable, and efficient fraud detection systems, considering their significant impact on financial security and the overall financial ecosystem. Full article
(This article belongs to the Special Issue New Advances in Cloud Computing and Its Latest Applications)
►▼ Show Figures

Figure 1

14 pages, 5030 KiB  
Article
A Linearized Open-Loop MDAC with Memory Effect Compensation Technique for High-Speed Pipelined ADC Stage
by Jie Wu, Qiao Meng, Shaocong Guo, Gaojing Li, Jianxun Shao and Sha Li
Electronics 2025, 14(9), 1753; https://doi.org/10.3390/electronics14091753 - 25 Apr 2025
Abstract
This paper presents a prototype open-loop pipelined stage in a 45 nm CMOS process for supporting 1.8 GS/s and 10-bit design specifications of pipelined ADCs. In order to alleviate the severe non-linearity expressed by open-loop MDACs, an innovative current-mode harmonic compensation is proposed [...] Read more.
This paper presents a prototype open-loop pipelined stage in a 45 nm CMOS process for supporting 1.8 GS/s and 10-bit design specifications of pipelined ADCs. In order to alleviate the severe non-linearity expressed by open-loop MDACs, an innovative current-mode harmonic compensation is proposed to provide input related third harmonic terms to cancel non-linearity. In addition, an effective double-sampling scheme is optimized by modifying compensation timing and input of a residual amplifier so that the pipelined stage can be immune to memory effect and improve power efficiency. The memory effect compensation scheme can provide a 21 dB improvement on output SNDR of the double-sampling pipelined stage. The simulation results illustrate that the open-loop pipelined ADC stage achieves an output SNDR of at least 52 dB with 840 mV input amplitude and 240 fF load while consuming only 11.24 mW. Full article
(This article belongs to the Section Microelectronics)
►▼ Show Figures

Figure 1

20 pages, 14942 KiB  
Article
Hybrid Energy Storage System for Regenerative Braking Utilization and Peak Power Decrease in 3 kV DC Railway Electrification System
by Adam Szeląg, Włodzimierz Jefimowski, Tadeusz Maciołek, Anatolii Nikitenko, Maciej Wieczorek and Mirosław Lewandowski
Electronics 2025, 14(9), 1752; https://doi.org/10.3390/electronics14091752 - 25 Apr 2025
Abstract
This paper proposes the sizing optimization method and energy management strategy for a stationary hybrid energy storage system dedicated to a DC traction power supply system. The hybrid energy storage system consists of two modules—a supercapacitor, mainly dedicated to regenerative energy utilization, and [...] Read more.
This paper proposes the sizing optimization method and energy management strategy for a stationary hybrid energy storage system dedicated to a DC traction power supply system. The hybrid energy storage system consists of two modules—a supercapacitor, mainly dedicated to regenerative energy utilization, and a Li-ion battery, aimed to peak power reduction. The sizing method and energy management strategy proposed in this paper aim to reduce the aging effect of lithium-ion batteries. It is shown that the parameters of both modules could be sized independently. The supercapacitor module parameters are sized based on the results of a simulation determining the regenerative power, resulting in limited catenary receptivity. The simulation model of the DC electrification system is validated by comparing the results of the simulation with the measurements of 15 min average power in a 24 h cycle as average values of one year. The battery module is sized based on the statistical data of 15 min substation power value occurrences. The battery energy capacity, its maximum discharge C-rate, and the conditions determining its operation are optimized to achieve the maximum ratio of annual income resulting from peak power reduction to annual operating cost resulting from the battery aging process and total life cycle. The case study prepared for a typical 3 kV DC substation with mixed railway traffic shows that peak power could be reduced by ~1 MW, giving a ~10-year payback period for battery module installation, while the energy consumption could be decreased by 1.9 MWh/24 h, giving a ~7.5-year payback period for supercapacitor module installation. The payback period of the whole energy storage system (ESS) is ~8.4 years. Full article
(This article belongs to the Special Issue Railway Traction Power Supply, 2nd Edition)
►▼ Show Figures

Figure 1

16 pages, 878 KiB  
Article
Distributed Adaptive Formation Control for Second-Order Multi-Agent Systems Without Collisions
by Juan Francisco Flores-Resendiz, Jesus David Aviles-Velazquez, Claudia Marquez, Rigoberto Martinez-Clark and Maria Alejandra Rojas-Ruiz
Electronics 2025, 14(9), 1751; https://doi.org/10.3390/electronics14091751 - 25 Apr 2025
Abstract
This paper presents an adaptive strategy to solve the formation control problem for a set of second-order agents with parametric uncertainty and nonlinearity. The strategy regards a group of agents where the nonlinearities and uncertainties are represented by a linearly parametrized term, which [...] Read more.
This paper presents an adaptive strategy to solve the formation control problem for a set of second-order agents with parametric uncertainty and nonlinearity. The strategy regards a group of agents where the nonlinearities and uncertainties are represented by a linearly parametrized term, which allows us to consider non-identical agents. In order to ensure the collision-free motion of agents, we propose the use of a repulsive vector field component that is applied only when a pair of agents becomes nearer than a predefined minimum bound. Numerical simulations were carried out to show the effectiveness of the proposed scheme. First, a simplified example was used to verify the key features of the control law, followed by a general case to illustrate the performance of the algorithm in a more complex scenario. Full article
(This article belongs to the Special Issue Research on Cooperative Control of Multi-agent Unmanned Systems)
►▼ Show Figures

Figure 1

40 pages, 4320 KiB  
Review
Federated Learning in Smart Healthcare: A Survey of Applications, Challenges, and Future Directions
by Mohammad Nasajpour, Seyedamin Pouriyeh, Reza M. Parizi, Meng Han, Fatemeh Mosaiyebzadeh, Liyuan Liu, Yixin Xie and Daniel Macêdo Batista
Electronics 2025, 14(9), 1750; https://doi.org/10.3390/electronics14091750 - 25 Apr 2025
Abstract
In recent years, novel technologies in smart healthcare systems have opened significant opportunities for diagnosis and treatment across various medical fields. Federated Learning (FL), a decentralized machine learning approach, trains shared models using local data from devices like wearables and hospital systems without [...] Read more.
In recent years, novel technologies in smart healthcare systems have opened significant opportunities for diagnosis and treatment across various medical fields. Federated Learning (FL), a decentralized machine learning approach, trains shared models using local data from devices like wearables and hospital systems without transferring sensitive information, offering a promising solution to privacy challenges in areas such as cancer prediction, COVID-19 detection, drug discovery, and medical image processing. This literature survey reviews FL architectures (e.g., FedHealth, PerFit), applications, and recent advancements, demonstrating their impact on healthcare through enhanced predictive models for patient care. Key findings include improved accuracy in wearable-based diagnostics and secure multi-institutional collaboration, though limitations persist. We also highlight open challenges, such as security risks, communication costs, and data heterogeneity, which require further research attention. Full article
►▼ Show Figures

Figure 1

20 pages, 5457 KiB  
Article
A Mathematical Method of Current-Carrying Capacity for Shore Power Cables in Port Microgrids
by Fei You, Mohd Abdul Talib Mat Yusoh, Nik Hakimi Nik Ali and Hao Yang
Electronics 2025, 14(9), 1749; https://doi.org/10.3390/electronics14091749 - 25 Apr 2025
Abstract
It is difficult to consider factors such as wind speed, water flow velocity, and solar radiation when using the IEC 60287 standard to calculate the current-carrying capacity of shore power cables in port microgrids. Therefore, based on the equivalent thermal circuit model and [...] Read more.
It is difficult to consider factors such as wind speed, water flow velocity, and solar radiation when using the IEC 60287 standard to calculate the current-carrying capacity of shore power cables in port microgrids. Therefore, based on the equivalent thermal circuit model and heat balance equation, this research takes solar radiation as the heat source of the cable used in port microgrids and proposes a mathematical calculation method for the current-carrying capacity of shore power cables based on the Newton–Raphson method. The influence of wind and water speed, environmental temperature, and solar radiation on current-carrying capacity is compared and analyzed using this mathematical calculation method and simulation calculation method. Shore power cables exhibit higher ampacity in water than air due to water’s superior thermal conductivity. Maximum ampacity difference occurs at 0.17 m/s flow (26.8 A analytically) and 0.066 m/s flow (64.4 A simulation). Air-laid cables show amplified ambient temperature effects from solar radiation, while water-laid cables demonstrate near-linear ampacity variations (Δ40 °C: 0–40 °C temperature range). This research can provide a reference for the revision of the standard for calculating the current-carrying capacity of shore power cables and optimizing renewable-energy-integrated port power systems. Full article
(This article belongs to the Special Issue Real-Time Monitoring and Intelligent Control for a Microgrid)
►▼ Show Figures

Figure 1

17 pages, 972 KiB  
Article
ITS-Rec: A Sequential Recommendation Model Using Item Textual Information
by Dongsoo Jang, Seok-Kee Lee and Qinglong Li
Electronics 2025, 14(9), 1748; https://doi.org/10.3390/electronics14091748 - 25 Apr 2025
Abstract
As the e-commerce industry rapidly expands, the number of users and items continues to grow, making it increasingly difficult to capture users’ purchasing patterns. Sequential recommendation models have emerged to address this issue by predicting the next item that a user is likely [...] Read more.
As the e-commerce industry rapidly expands, the number of users and items continues to grow, making it increasingly difficult to capture users’ purchasing patterns. Sequential recommendation models have emerged to address this issue by predicting the next item that a user is likely to purchase based on their historical behavior. However, most previous studies have focused primarily on modeling item sequences using item IDs without leveraging rich item-level information. To address this limitation, we propose a sequential recommendation model called ITS-Rec that incorporates various types of textual item information, including item titles, descriptions, and online reviews. By integrating these components into item representations, the model captures both detailed item characteristics and signals related to purchasing motivation. ITS-Rec is built on a self-attention-based architecture that enables the model to effectively learn both the long- and short-term user preferences. Experiments were conducted using real-world Amazon.com data, and the proposed model was compared to several state-of-the-art sequential recommendation models. The results demonstrate that ITS-Rec significantly outperforms the baseline models in terms of Hit Ratio (HR) and Normalized Discounted Cumulative Gain (NDCG). Further analysis showed that online reviews contributed the most to performance gains among textual components. This study highlights the value of incorporating textual features into sequential recommendations and provides practical insights into enhancing recommendation performance through richer item representations. Full article
(This article belongs to the Special Issue Data-Driven AI Approaches with Applications in Social Network, Media Analytics and Smart Cities)
►▼ Show Figures

Figure 1

20 pages, 4930 KiB  
Article
Light Field Super-Resolution via Dual-Domain High-Frequency Restoration and State-Space Fusion
by Zhineng Zhang, Tao Yan, Hao Huang, Jinsheng Liu, Chenglong Wang and Cihang Wei
Electronics 2025, 14(9), 1747; https://doi.org/10.3390/electronics14091747 - 25 Apr 2025
Abstract
The current light field super-resolution methods mainly face the following challenges: difficulty in handling redundant information in light fields; heavy reliance on the spatial domain to recover details; and insufficient interaction of spatial and angular features. We propose a novel light field super-resolution [...] Read more.
The current light field super-resolution methods mainly face the following challenges: difficulty in handling redundant information in light fields; heavy reliance on the spatial domain to recover details; and insufficient interaction of spatial and angular features. We propose a novel light field super-resolution (LF-SR) network, termed DHSFNet, which effectively enhances super-resolution performance from a dual-domain perspective, encompassing both the frequency and spatial domains. Our DHSFNet contains three key points. (1) A local sparse angular attention module (LSAA) is proposed to selectively capture relationships between adjacent sub-views using geometric prior information to reduce computational complexity. (2) We design a dual-domain high-frequency restoration sub-network, with a frequency-domain branch using mask-guided multi-scale discrete cosine transform (DCT) restoration and a spatial-domain branch employing multi-scale cross-attention to recover texture details. (3) A Mamba-based fusion module (MF) is introduced to efficiently facilitate global spatial–angular interaction, which achieves linear complexity and outperforms Transformer-based methods in both accuracy and speed. Comprehensive experiments conducted on three benchmark datasets demonstrate the superior performance of our method in the LF-SR task. Full article
►▼ Show Figures

Figure 1

21 pages, 13080 KiB  
Article
Color Normalization Through a Simulated Color Checker Using Generative Adversarial Networks
by Albert Siré Langa, Ramón Reig Bolaño, Sergi Grau Carrión and Ibon Uribe Elorrieta
Electronics 2025, 14(9), 1746; https://doi.org/10.3390/electronics14091746 - 25 Apr 2025
Abstract
Digital cameras often struggle to reproduce the true colors perceived by the human eye due to lighting geometry and illuminant color. This research proposes an innovative approach for color normalization in digital photographs. A machine learning algorithm combined with an external physical color [...] Read more.
Digital cameras often struggle to reproduce the true colors perceived by the human eye due to lighting geometry and illuminant color. This research proposes an innovative approach for color normalization in digital photographs. A machine learning algorithm combined with an external physical color checker achieves color normalization. To address the limitations of relying on a physical color checker, our approach employs a generative adversarial network capable of replicating the color normalization process without the need for a physical reference. This network (GAN-CN-CC) incorporates a custom loss function specifically designed to minimize errors in color generation. The proposed algorithm yields the lowest coefficient of variation in the normalized median intensity (NMI), while maintaining a standard deviation comparable to that of conventional methods such as Gray World and Max-RGB. The algorithm eliminates the need for a color checker in color normalization, making it more practical in scenarios where inclusion of the checker is challenging. The proposed method has been fine-tuned and validated, demonstrating high effectiveness and adaptability. Full article
(This article belongs to the Special Issue Machine Learning in Data Analytics and Prediction)
►▼ Show Figures

Figure 1

21 pages, 5674 KiB  
Article
Reality Head-Up Display Navigation Design in Extreme Weather Conditions: Enhancing Driving Experience in Rain and Fog
by Qi Zhu and Ziqi Liu
Electronics 2025, 14(9), 1745; https://doi.org/10.3390/electronics14091745 - 25 Apr 2025
Abstract
This study investigates the impact of extreme weather conditions (specifically heavy rain and fog) on drivers’ situational awareness by analyzing variations in illumination levels. The primary objective is to identify optimal color wavelengths for low-light environments, thereby providing a theoretical foundation for the [...] Read more.
This study investigates the impact of extreme weather conditions (specifically heavy rain and fog) on drivers’ situational awareness by analyzing variations in illumination levels. The primary objective is to identify optimal color wavelengths for low-light environments, thereby providing a theoretical foundation for the design of augmented reality head-up display in adverse weather conditions. A within-subjects experimental design was employed with 26 participants in a simulated driving environment. Participants were exposed to different illumination levels and AR-HUD colors. Eye-tracking metrics, including fixation duration, visit duration, and fixation count, were recorded alongside situational awareness ratings to assess cognitive load and information processing efficiency. The results revealed that the yellow AR-HUD significantly enhanced situational awareness and reduced cognitive load in foggy conditions. While subjective assessments indicated no substantial effect of lighting conditions, objective measurements demonstrated the superior effectiveness of the yellow AR-HUD under foggy weather. These findings suggest that yellow AR-HUD navigation icons are more suitable for extreme weather environments, offering potential improvements in driving performance and overall road safety. Full article
(This article belongs to the Special Issue Advancements in Sensing and Perception for Autonomous Vehicles in Adverse Environmental Conditions)
►▼ Show Figures

Figure 1

18 pages, 1008 KiB  
Article
LLM-Based Query Expansion with Gaussian Kernel Semantic Enhancement for Dense Retrieval
by Min Pan, Wenrui Xiong, Shuting Zhou, Mengfei Gao and Jinguang Chen
Electronics 2025, 14(9), 1744; https://doi.org/10.3390/electronics14091744 - 24 Apr 2025
Abstract
In the field of Information Retrieval (IR), user-submitted keyword queries often fail to accurately represent users’ true search intent. With the rapid advancement of artificial intelligence, particularly in natural language processing (NLP), query expansion (QE) based on large language models (LLMs) has emerged [...] Read more.
In the field of Information Retrieval (IR), user-submitted keyword queries often fail to accurately represent users’ true search intent. With the rapid advancement of artificial intelligence, particularly in natural language processing (NLP), query expansion (QE) based on large language models (LLMs) has emerged as a key strategy for improving retrieval effectiveness. However, such methods often introduce query topic drift, which negatively impacts retrieval accuracy and efficiency. To address this issue, this study proposes an LLM-based QE framework that incorporates a Gaussian kernel-enhanced semantic space for dense retrieval. Specifically, the model first employs LLMs to expand the semantic dimensions of the initial query, generating multiple query representations. Then, by introducing a Gaussian kernel semantic space, it captures deep semantic relationships among these query vectors, refining their semantic distribution to better represent the original query’s intent. Finally, the ColBERTv2 model is utilized to retrieve documents based on the enhanced query representations, enabling precise relevance assessment and improving retrieval performance. To validate the effectiveness of the proposed approach, extensive empirical evaluations were conducted on the MS MARCO passage ranking dataset. The model was systematically assessed using key metrics, including MAP, NDCG@10, MRR@10, and Recall@1000. Experimental results demonstrate that the proposed method outperforms existing approaches across multiple metrics, significantly improving retrieval precision while effectively mitigating query drift, offering a novel approach for building efficient QE mechanisms. Full article
(This article belongs to the Special Issue Innovative Applications of Large Language Models in Natural Language Processing (NLP))
►▼ Show Figures

Figure 1

23 pages, 3096 KiB  
Review
A Critical Review of Techniques for the Experimental Extraction of the Thermal Resistance of Bipolar Transistors from DC Measurements—Part II: Approaches Based on Intersection Points
by Vincenzo d’Alessandro, Antonio Pio Catalano and Ciro Scognamillo
Electronics 2025, 14(9), 1743; https://doi.org/10.3390/electronics14091743 - 24 Apr 2025
Abstract
This work constitutes Part II of a comprehensive three-part study critically reviewing techniques for the indirect extraction of the thermal resistance in bipolar transistors using simple DC current/voltage measurements. While Part I focused on thermometer-based methods, this study examines techniques that rely on [...] Read more.
This work constitutes Part II of a comprehensive three-part study critically reviewing techniques for the indirect extraction of the thermal resistance in bipolar transistors using simple DC current/voltage measurements. While Part I focused on thermometer-based methods, this study examines techniques that rely on intersection points between electrical characteristics. The accuracy of these methods is assessed by applying them to DC curves obtained through PSPICE simulations of an in-house transistor model incorporating nonlinear thermal effects, and comparing the extracted thermal resistance data with the thermal resistance formulation embedded in the model. An InGaP/GaAs HBT and a Si/SiGe HBT for high-frequency applications are considered as case-studies. The analysis highlights key drawbacks affecting the methods, including theoretical approximations and sensitivity to the selection of intersection points. Among the techniques examined, only one adequately accounts for the nonlinear thermal behavior, though its original formulation presents practical limitations. To tackle this problem, we propose an improved and refined version of the approach that offers enhanced accuracy at the cost of increased complexity. Full article
►▼ Show Figures

Figure 1

28 pages, 2001 KiB  
Review
Recent Research Progress of Graph Neural Networks in Computer Vision
by Zhiyong Jia, Chuang Wang, Yang Wang, Xinrui Gao, Bingtao Li, Lifeng Yin and Huayue Chen
Electronics 2025, 14(9), 1742; https://doi.org/10.3390/electronics14091742 - 24 Apr 2025
Abstract
Graph neural networks (GNNs) have demonstrated significant potential in the field of computer vision in recent years, particularly in handling non-Euclidean data and capturing complex spatial and semantic relationships. This paper provides a comprehensive review of the latest research on GNNs in computer [...] Read more.
Graph neural networks (GNNs) have demonstrated significant potential in the field of computer vision in recent years, particularly in handling non-Euclidean data and capturing complex spatial and semantic relationships. This paper provides a comprehensive review of the latest research on GNNs in computer vision, with a focus on their applications in image processing, video analysis, and multimodal data fusion. First, we briefly introduce common GNN models, such as graph convolutional networks (GCN) and graph attention networks (GAT), and analyze their advantages in image and video data processing. Subsequently, this paper delves into the applications of GNNs in tasks such as object detection, image segmentation, and video action recognition, particularly in capturing inter-region dependencies and spatiotemporal dynamics. Finally, the paper discusses the applications of GNNs in multimodal data fusion tasks such as image–text matching and cross-modal retrieval, and highlights the main challenges faced by GNNs in computer vision, including computational complexity, dynamic graph modeling, heterogeneous graph processing, and interpretability issues. This paper provides a comprehensive understanding of the applications of GNNs in computer vision for both academia and industry and envisions future research directions. Full article
(This article belongs to the Special Issue AI Synergy: Vision, Language, and Modality)
►▼ Show Figures

Figure 1

18 pages, 1897 KiB  
Article
Multi-Path Convolutional Architecture with Channel-Wise Attention for Multiclass Brain Tumor Detection in Magnetic Resonance Imaging Scans
by Muneeb A. Khan, Tsagaanchuluun Sugir, Byambaa Dorj, Ganchimeg Uuganchimeg, Seonuck Paek, Khurelbaatar Zagarzusem and Heemin Park
Electronics 2025, 14(9), 1741; https://doi.org/10.3390/electronics14091741 - 24 Apr 2025
Abstract
Accurately detecting and classifying brain tumors in magnetic resonance imaging (MRI) scans poses formidable challenges, stemming from the heterogeneous presentation of tumors and the need for reliable, real-time diagnostic outputs. In this paper, we propose a novel multi-path convolutional architecture enhanced with channel-wise [...] Read more.
Accurately detecting and classifying brain tumors in magnetic resonance imaging (MRI) scans poses formidable challenges, stemming from the heterogeneous presentation of tumors and the need for reliable, real-time diagnostic outputs. In this paper, we propose a novel multi-path convolutional architecture enhanced with channel-wise attention mechanisms, evaluated on a comprehensive four-class brain tumor dataset. Specifically: (i) we design a parallel feature extraction strategy that captures nuanced tumor morphologies, while channel-wise attention refines salient characteristics; (ii) we employ systematic data augmentation, yielding a balanced dataset of 6380 MRI scans to bolster model generalization; (iii) we compare the proposed architecture against state-of-the-art models, demonstrating superior diagnostic performance with 97.52% accuracy, 97.63% precision, 97.18% recall, 98.32% specificity, and an F1-score of 97.36%; and (iv) we report an inference speed of 5.13 ms per scan, alongside a higher memory footprint of approximately 26 GB, underscoring both the feasibility for real-time clinical application and the importance of resource considerations. These findings collectively highlight the proposed framework’s potential for improving automated brain tumor detection workflows and prompt further optimization for broader clinical deployment. Full article
(This article belongs to the Special Issue Revolutionizing Medical Image Analysis with Deep Learning, 2nd Edition)
►▼ Show Figures

Figure 1

19 pages, 4018 KiB  
Article
Research on Weather Recognition Based on a Field Programmable Gate Array and Lightweight Convolutional Neural Network
by Liying Chen, Fan Luo, Fei Wang and Liangfu Lv
Electronics 2025, 14(9), 1740; https://doi.org/10.3390/electronics14091740 - 24 Apr 2025
Abstract
With the rapid development of deep learning, weather recognition has become a research hotspot in the field of computer vision, and the research on field programmable gate array (FPGA) acceleration based on deep learning algorithms has received more and more attention, based on [...] Read more.
With the rapid development of deep learning, weather recognition has become a research hotspot in the field of computer vision, and the research on field programmable gate array (FPGA) acceleration based on deep learning algorithms has received more and more attention, based on which, we propose a method to implement deep neural networks for weather recognition in a small-scale FPGA. First, we train a deep separable convolutional neural network model for weather recognition to reduce the parameters and speed up the performance of hardware implementation. However, large-scale computation also brings the problem of excessive power consumption, which greatly limits the deployment of high-performance network models on mobile platforms. Therefore, we use a lightweight convolutional neural network approach to reduce the scale of computation, and the main idea of lightweight is to use fewer bits to store the weights. In addition, a hardware implementation of this model is proposed to speed up the operation and save on-chip resource consumption. Finally, the network model is deployed on a Xilinx ZYNQ xc7z020 FPGA to verify the accuracy of the recognition results, and the accelerated solution succeeds in achieving excellent performance with a speed of 108 FPS and 3.256 W of power consumption. The purpose of this design is to be able to accurately recognize the weather and deliver current environmental weather information to UAV (unmanned aerial vehicle) pilots and other staff who need to consider the weather, so that they can accurately grasp the current environmental weather conditions at any time. When the weather conditions change, the information can be obtained in a timely and effective manner to make the correct judgment, to ensure the flight of the UAV, and to avoid the equipment being affected by the weather leading to equipment damage and failure of the flight mission. With the help of this design, the UAV flight mission can be better completed. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Figure 1

20 pages, 5808 KiB  
Article
Enhanced YOLOv7 Based on Channel Attention Mechanism for Nearshore Ship Detection
by Qingyun Zhu, Zhen Zhang and Ruizhe Mu
Electronics 2025, 14(9), 1739; https://doi.org/10.3390/electronics14091739 - 24 Apr 2025
Abstract
Nearshore ship detection is an important task in marine monitoring, playing a significant role in navigation safety and controlling illegal smuggling. The continuous research and development of Synthetic Aperture Radar (SAR) technology is not only of great importance in military and maritime security [...] Read more.
Nearshore ship detection is an important task in marine monitoring, playing a significant role in navigation safety and controlling illegal smuggling. The continuous research and development of Synthetic Aperture Radar (SAR) technology is not only of great importance in military and maritime security fields but also has great potential in civilian fields, such as disaster emergency response, marine resource monitoring, and environmental protection. Due to the limited sample size of nearshore ship datasets, it is difficult to meet the demand for the large quantity of training data required by existing deep learning algorithms, which limits the recognition accuracy. At the same time, artificial environmental features such as buildings can cause significant interference to SAR imaging, making it more difficult to distinguish ships from the background. Ship target images are greatly affected by speckle noise, posing additional challenges to data-driven recognition methods. Therefore, we utilized a Concurrent Single-Image GAN (ConSinGAN) to generate high-quality synthetic samples for re-labeling and fused them with the dataset extracted from the SAR-Ship dataset for nearshore image extraction and dataset division. Experimental analysis showed that the ship recognition model trained with augmented images had an accuracy increase of 4.66%, a recall rate increase of 3.68%, and an average precision (AP) with Intersection over Union (IoU) at 0.5 increased by 3.24%. Subsequently, an enhanced YOLOv7 algorithm (YOLOv7 + ESE) incorporating channel-wise information fusion was developed based on the YOLOv7 architecture integrated with the Squeeze-and-Excitation (SE) channel attention mechanism. Through comparative experiments, the analytical results demonstrated that the proposed algorithm achieved performance improvements of 0.36% in precision, 0.52% in recall, and 0.65% in average precision (AP@0.5) compared to the baseline model. This optimized architecture enables accurate detection of nearshore ship targets in SAR imagery. Full article
(This article belongs to the Special Issue Intelligent Systems in Industry 4.0)
►▼ Show Figures

Figure 1

25 pages, 5122 KiB  
Article
Detection of Exoplanets in Transit Light Curves with Conditional Flow Matching and XGBoost
by Stefano Fiscale, Alessio Ferone, Angelo Ciaramella, Laura Inno, Massimiliano Giordano Orsini, Giovanni Covone and Alessandra Rotundi
Electronics 2025, 14(9), 1738; https://doi.org/10.3390/electronics14091738 - 24 Apr 2025
Abstract
NASA’s space-based telescopes Kepler and Transiting Exoplanet Survey Satellite (TESS) have detected billions of potential planetary signatures, typically classified with Convolutional Neural Networks (CNNs). In this study, we introduce a hybrid model that combines deep learning, dimensionality reduction, decision trees, and diffusion models [...] Read more.
NASA’s space-based telescopes Kepler and Transiting Exoplanet Survey Satellite (TESS) have detected billions of potential planetary signatures, typically classified with Convolutional Neural Networks (CNNs). In this study, we introduce a hybrid model that combines deep learning, dimensionality reduction, decision trees, and diffusion models to distinguish planetary transits from astrophysical false positives and instrumental artifacts. Our model consists of three main components: (i) feature extraction using the CNN VGG19, (ii) dimensionality reduction through t-Distributed Stochastic Neighbor Embedding (t-SNE), and (iii) classification using Conditional Flow Matching (CFM) and XGBoost. We evaluated the model on two Kepler and one TESS datasets, achieving F1-scores of 98% and 100%, respectively. Our results demonstrate the effectiveness of VGG19 in extracting discriminative patterns from data, t-SNE in projecting features in a lower dimensional space where they can be most effectively classified, and CFM with XGBoost in enabling robust classification with minimal computational cost. This study highlights that a hybrid approach leveraging deep learning and dimensionality reduction allows one to achieve state-of-the-art performance in exoplanet detection while maintaining a low computational cost. Future work will explore the use of adaptive dimensionality reduction methods and the application to data from upcoming missions like the ESA’s PLATO mission. Full article
(This article belongs to the Special Issue Image Processing Based on Convolution Neural Network: 2nd Edition)
►▼ Show Figures

Figure 1

11 pages, 1677 KiB  
Article
A Novel Darlington Structure Power Switch Using a Vacuum Field Emission Transistor
by Yulong Ding, Yanlin Ke, Juncong She, Yu Zhang and Shaozhi Deng
Electronics 2025, 14(9), 1737; https://doi.org/10.3390/electronics14091737 - 24 Apr 2025
Abstract
This study proposes a power switch combining a vacuum field emission transistor (VFET) as a controlled transistor with a power bipolar Darlington transistor (DT) as an output transistor, termed the VFET–DT structure. Compared to the MOS–bipolar Darlington power switch, the VFET–DT structure achieves [...] Read more.
This study proposes a power switch combining a vacuum field emission transistor (VFET) as a controlled transistor with a power bipolar Darlington transistor (DT) as an output transistor, termed the VFET–DT structure. Compared to the MOS–bipolar Darlington power switch, the VFET–DT structure achieves an extremely low off-state leakage current and high-voltage withstanding capability due to the field emission mechanism of the VFET. It can also avoid the Miller effect that results from incorporating the load resistance into the feedback loop. The high gain and high-power capacity can be achieved due to the cascade of DT. The device’s typical electrical characteristics were theoretically investigated by simulation. The VFET–DT structure exhibited a high-power capacity of 20 A and 400 V with a minimum conduction voltage drop of 1.316 V and a switching frequency of 100 kHz. The results demonstrated that the combination of a vacuum transistor and a solid-state transistor combines the advantages of both and benefits the performance of the power switch. Full article
(This article belongs to the Special Issue Vacuum Electronics: From Micro to Nano)
►▼ Show Figures

Figure 1

14 pages, 287 KiB  
Article
Robust Face Recognition Based on the Wing Loss and the ℓ1 Penalty
by Yaoyao Yun and Jianwen Xu
Electronics 2025, 14(9), 1736; https://doi.org/10.3390/electronics14091736 - 24 Apr 2025
Abstract
In recent years, face recognition under occluded or corrupted conditions has emerged as a prominent research topic. The advancement in sparse sampling techniques based on regression analysis has provided a novel solution to this challenge. Currently, numerous regression-based sparse sampling models have been [...] Read more.
In recent years, face recognition under occluded or corrupted conditions has emerged as a prominent research topic. The advancement in sparse sampling techniques based on regression analysis has provided a novel solution to this challenge. Currently, numerous regression-based sparse sampling models have been investigated by researchers to address this problem. However, the recognition accuracy of most existing models deteriorates significantly when handling heavily occluded or severely corrupted facial images. To overcome this limitation, this paper proposes a wing-constrained sparse coding (WCSC) model and its weighted variant (weighted wing-constrained sparse coding, WWCSC) for robust face recognition in complex scenarios. The corresponding minimization problems are solved using the alternating direction method of multipliers (ADMM) algorithm. Extensive experiments are conducted on four benchmark face databases: the Olivetti Research Laboratory (ORL) database, the Yale database, the AR database and the Face Recognition Technology (FERET) database, to evaluate the proposed method’s performance. Comparative results demonstrate that the WWCSC model maintains superior recognition rates even under challenging conditions involving significant occlusion or corruption, highlighting its remarkable robustness in face recognition tasks. This study provides both theoretical and empirical validation for the effectiveness of the proposed approach. Full article
►▼ Show Figures

Figure 1

22 pages, 46829 KiB  
Article
Waveshift 2.0: An Improved Physics-Driven Data Augmentation Strategy in Fine-Grained Image Classification
by Gent Imeraj and Hitoshi Iyatomi
Electronics 2025, 14(9), 1735; https://doi.org/10.3390/electronics14091735 - 24 Apr 2025
Abstract
This paper presents Waveshift Augmentation 2.0 (WS 2.0), an enhanced version of the previously proposed Waveshift Augmentation (WS 1.0), a novel data augmentation technique inspired by light propagation dynamics in optical systems. While WS 1.0 introduced phase-based wavefront transformations under the assumption of [...] Read more.
This paper presents Waveshift Augmentation 2.0 (WS 2.0), an enhanced version of the previously proposed Waveshift Augmentation (WS 1.0), a novel data augmentation technique inspired by light propagation dynamics in optical systems. While WS 1.0 introduced phase-based wavefront transformations under the assumption of an infinitesimally small aperture, WS 2.0 incorporates an additional aperture-dependent hyperparameter that models real-world optical attenuation. This refinement enables broader frequency modulation and greater diversity in image transformations while preserving compatibility with well-established data augmentation pipelines such as CLAHE, AugMix, and RandAugment. Evaluated across a wide range of tasks, including medical imaging, fine-grained object recognition, and grayscale image classification, WS 2.0 consistently outperformed both WS 1.0 and standard geometric augmentation. Notably, when benchmarked against geometric augmentation alone, it achieved average macro-F1 improvements of +1.48 (EfficientNetV2), +0.65 (ConvNeXt), and +0.73 (Swin Transformer), with gains of up to +9.32 points in medical datasets. These results demonstrate that WS 2.0 advances physics-based augmentation by enhancing generalization without sacrificing modularity or preprocessing efficiency, offering a scalable and realistic augmentation strategy for complex imaging domains. Full article
(This article belongs to the Special Issue New Trends in Computer Vision and Image Processing)
►▼ Show Figures

Figure 1

25 pages, 9236 KiB  
Article
Enhancing Medium-Orbit Satellite Orbit Prediction: Application and Experimental Validation of the BiLSTM-TS Model
by Yang Guo, Bingchuan Li, Xueshu Shi, Zhengxu Zhao, Jian Sun and Jinsheng Wang
Electronics 2025, 14(9), 1734; https://doi.org/10.3390/electronics14091734 - 24 Apr 2025
Abstract
To mitigate the limited accuracy of the Simplified General Perturbations 4 (SGP4) model in predicting medium-orbit satellite trajectories, we propose an enhanced methodology integrating deep learning with traditional algorithms. The developed BiLSTM-TS forecasting framework comprises a Bidirectional Long Short-Term Memory (BiLSTM) network, trend [...] Read more.
To mitigate the limited accuracy of the Simplified General Perturbations 4 (SGP4) model in predicting medium-orbit satellite trajectories, we propose an enhanced methodology integrating deep learning with traditional algorithms. The developed BiLSTM-TS forecasting framework comprises a Bidirectional Long Short-Term Memory (BiLSTM) network, trend analysis module (T), and seasonal decomposition module (S). This architecture effectively captures sequential dependencies, trend variations, and periodic patterns within time series data, thereby improving prediction interpretability. In our experimental validation, we chose Beidou-2 M6 (C14), GSAT0203 (GALILEO 7), and the Global Positioning System (GPS) satellite named GPS BIIR-13 (PRN 02) as representative satellites. Satellite position data derived from conventional orbital models were input into the BiLSTM-TS framework for statistical learning to predict orbital deviations. These predicted errors were subsequently combined with SGP4 model outputs obtained through Two-Line Element set (TLE) data analysis to minimize overall trajectory inaccuracies. Using BeiDou-2 M6 (C14) as a case study, results indicated that the BiLSTM-TS implementation achieved significant error reduction; mean-squared error along the X-axis was reduced to 0.0309 km2, with mean absolute error of 0.1245 km, and maximum absolute error was constrained to 0.4448 km. Full article
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-70
Previous Issue
Back to TopTop