Computers & Electrical Engineering最新文献

{"title":"Developing LoRa-IoT infrastructure based advanced-airborne security systems for public healthcare centers using machine learning","authors":"Kanak Kumar ,&nbsp;Anshul Verma ,&nbsp;Pradeepika Verma","doi":"10.1016/j.compeleceng.2025.110703","DOIUrl":"10.1016/j.compeleceng.2025.110703","url":null,"abstract":"<div><div>Airborne security poses a growing threat in indoor and outdoor environments such as healthcare centres. The rapid growth of the Internet of Things (IoT) has integrated safety systems into our daily lives. This paper presents a solar-powered, energy-efficient and sustainable <strong>L</strong>ong Range (LoRa) technology enabled <strong>I</strong>oT infrastructure based <strong>A</strong>dvanced-<strong>A</strong>irborne <strong>S</strong>ecurity systems <strong>for P</strong>ublic <strong>H</strong>ealthcare <strong>C</strong>enters (LIA2S4PHC) by using Machine Learning (ML) for the detection and analysis of the signature of harmful pollutant/odors/volatile organic compounds (VOCs)/gases. For LIA2S4PHC framework development, we have applied two LoRa SX1278 (transmitter and receiver end), two ESP 32 microcontrollers, six tin-oxide-based cross-selective gas sensors, one particulate matter sensor, one temperature and humidity sensor, and one GPS module. LoRa is used to transmit airborne signatures from the experimental zone to the central monitoring station (CMS) for further processing. On performing the LIA2S4PHC range test, it was found that up to 745 m information was obtained correctly in open space. The messages started getting missed after that, and at about 765 m there was no reception. The captured airborne signatures were pre-processed using the standardized principal component analysis (SPCA) for 3D scattering. Naïve Bayes (NB), support vector machine (SVM), multi-layer perceptron (MLP), Categorical Boosting (CatBoost), and tabular network (TabNet) models were applied to classify the types of harmful gases/odors/airborne signatures with classification average accuracies such as 82%, 84%, 91%, 91%, and 92%, respectively. TabNet classification model outperforms other classification models with 92% accuracy. We found <span><math><mrow><mn>4</mn><mo>.</mo><mn>7</mn><mo>×</mo><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>6</mn></mrow></msup></mrow></math></span> and <span><math><mrow><mn>1</mn><mo>.</mo><mn>34</mn><mo>×</mo><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>3</mn></mrow></msup></mrow></math></span> minimum and maximum MSE respectively. LIA2S4PHC is a full y scalable and energy-efficient system applicable for public healthcare centres and a wide variety of applications. Our dataset and code are available at <strong>GitHub link</strong>: <span><span>https://github.com/kanakkumarcs24/Advanced-Airborne-Security-Systems-for-Public-Healthcare-Centers</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110703"},"PeriodicalIF":4.9,"publicationDate":"2025-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145057377","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quality-guaranteed steganographed-MECG signal compression using adaptive truncation of DCT and SVD coefficients and ASCII encoding","authors":"Sourav Kumar Mukhopadhyay ,&nbsp;Sridhar Krishnan","doi":"10.1016/j.compeleceng.2025.110704","DOIUrl":"10.1016/j.compeleceng.2025.110704","url":null,"abstract":"<div><div>With the proliferation of wearable healthcare devices and garments in the last decade, the necessity of the storage capacity of the acquired biomedical signals; particularly multi-lead electrocardiogram (MECG) signals, and the importance of securing users’ personal information have increased significantly. However, existing MECG compression and steganography algorithms are insufficient to address these challenges effectively. This paper presents a discrete cosine transform, singular value decomposition, and American standard code for information interchange (ASCII) character encoding-based highly efficient quality-guaranteed steganographed MECG compression algorithm. The algorithm is tested on three publicly available MECG databases totalling 2.98 months, and its performance is assessed through both qualitative and quantitative measures. The algorithm attains a compression ratio that is much higher than that provided by other algorithms that are developed to compress the MECG signals only. The benefits of using the proposed algorithm are fivefolds: first, the clinical qualities of the reconstructed MECG signals can be controlled precisely, second, user’s personal information is restored with no error, third, reconstruction error of the MECG signals is dependent neither on the size of the user’s information nor on the steganography operation, fourth, the probability of guesstimating the security-key is close to zero, and fifth, high compression performance.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110704"},"PeriodicalIF":4.9,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Crime prediction against women in surveillance videos using deep learning models: A review","authors":"Kanchan Ganesh Dhuri ,&nbsp;Sunita Patil","doi":"10.1016/j.compeleceng.2025.110651","DOIUrl":"10.1016/j.compeleceng.2025.110651","url":null,"abstract":"<div><div>In recent years, violence against women has gained prominence in Indian debate. Ensuring secure public transportation environments is crucial for enhancing women’s active engagement in the economic, educational and social domains. Although various studies have been undertaken on the issue in recent years, the majority of present research lacks substantial in-depth analysis. Furthermore, a notable shortcoming found in these studies is the insufficient attention paid to the possibility of cyber-attacks on security video systems. This study provides a detailed overview of recent advances in deep learning-based models for identifying antisocial conduct against women in surveillance films. It assesses cutting-edge approaches and publicly available audio-visual datasets used in model training, with a particular emphasis on their relevance to real-world transportation scenarios. The study investigates the function of intelligent video surveillance systems and the integration of multi-modal data in crime detection. Important issues are noted, including model generalizability, dataset constraints and ethical considerations. The results offer crucial insights into the design of reliable, automated safety systems, making them a fundamental resource for academics and system developers. This study impacts improving public safety and lowering gender-based violence by providing guidance for the creation of scalable and effective crime detection mechanisms. It establishes the foundation for upcoming advancements in deep learning used in public transit security and urban surveillance.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110651"},"PeriodicalIF":4.9,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analytical approach for inductance calculation of a rectangular curved coil for wireless charging of an autonomous underwater vehicle","authors":"Dilruba Ferdous ,&nbsp;Suman Saha ,&nbsp;Ranjit Ray","doi":"10.1016/j.compeleceng.2025.110706","DOIUrl":"10.1016/j.compeleceng.2025.110706","url":null,"abstract":"<div><div>Coil geometry serves a crucial role in design of wireless power transfer (WPT) systems for autonomous underwater vehicles (AUVs). Conventional coils designs exhibit shortcomings associated with its compatibility with AUV hull structure. Addressing to this, a rectangular curved coil has been proposed in this work. Analytical estimation of self-inductance (SI) and mutual inductance (MI) is necessary for coil design prior to time-consuming simulations or expensive physical construction. However, the analytical approach for conventional coil geometrics are not applicable for a rectangular curved coil. Hence, a mathematical model for estimating SI and MI of a rectangular curved coil is derived and established here. Furthermore, the proposed analytical model is evaluated through Ansys simulation and experimentation carried out on a developed laboratory scale prototype of the rectangular curved coil. Apart from that, the variations in MI at different distances and misalignment conditions of the proposed coil has been performed in Ansys platform.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110706"},"PeriodicalIF":4.9,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on wind power prediction with secondary decomposition and multi-algorithm fusion for complex nonlinear time series","authors":"Qizhen Jia ,&nbsp;Kai Kang ,&nbsp;Beier Wang ,&nbsp;Yuehao Wu ,&nbsp;Yibo Zhang ,&nbsp;Fusen Guo","doi":"10.1016/j.compeleceng.2025.110688","DOIUrl":"10.1016/j.compeleceng.2025.110688","url":null,"abstract":"<div><div>Wind power forecasting plays a critical role in maintaining grid stability and enhancing electricity market competitiveness. However, the inherent nonlinearity and noise in wind power data—driven by multiple influencing factors such as wind speed and temperature—pose significant challenges for accurate prediction. To address this, we propose a hybrid forecasting framework that integrates secondary decomposition with multiple predictive algorithms. First, Singular Spectrum Analysis (SSA) is employed to extract the primary trend and periodic components, followed by Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) for further signal decomposition. Fuzzy entropy is then used to reconstruct the components into high-frequency, mid-frequency, and trend sub-series, effectively denoising the data and capturing multiscale features. For forecasting, an IDF-Informer model—incorporating Intra-layer Directional Fusion and optimized via a multi-strategy Improved Sand Cat Swarm Optimization (ISCSO)—is used for the high- and mid-frequency components, while the trend component is predicted using a Random Vector Functional Link Network without Direct(RVFLwoDL). Experimental results across the four seasons show that the proposed method achieves consistent improvements in RMSE, MAE, MAPE, and <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> compared to other models.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110688"},"PeriodicalIF":4.9,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-source Feature Map Distillation for enhanced low-resolution object recognition","authors":"Jin Ren ,&nbsp;Qinling Zhou ,&nbsp;Shunzhi Yang ,&nbsp;Jinfeng Yang","doi":"10.1016/j.compeleceng.2025.110710","DOIUrl":"10.1016/j.compeleceng.2025.110710","url":null,"abstract":"<div><div>Knowledge distillation is an effective method for addressing the problem of low-resolution object recognition. However, due to resolution differences, the feature map sizes become inconsistent, making it difficult for the student model to fully learn the rich privileged information contained in the teacher model. Our previous work addressed this issue through a feature decoder, achieving cross-resolution feature map distillation. However, it fails to fully leverage both high-resolution samples and their feature maps to extract more privileged information at the distillation point. To this end, this paper proposes a Multi-source Feature Map Distillation (MsFMD) method to further improve the performance of low-resolution object recognition in practical applications such as intelligent video surveillance. We design a feature decoder with a channel attention mechanism to better leverage the privileged information from the teacher model and employ multi-level decoder modules to process deep features, achieving multi-level feature map distillation. Additionally, this paper introduces a joint data augmentation method, effectively enhancing the student model’s adaptability and robustness across samples with varying resolutions. The overall performance of MsFMD is verified in multiple recognition tasks by comparing it with state-of-the-art knowledge distillation methods on low-resolution and noisy objects.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110710"},"PeriodicalIF":4.9,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A high-efficiency wide dynamic-range rectifier (0.2–0.9 GHz) for wireless power transfer in biomedical implants","authors":"Marwa Mansour ,&nbsp;Mohamed Aboualalaa ,&nbsp;Islam Mansour","doi":"10.1016/j.compeleceng.2025.110708","DOIUrl":"10.1016/j.compeleceng.2025.110708","url":null,"abstract":"<div><div>In this article, a new wide-band, large dynamic-range (DR), high-efficiency rectifier is utilized in Wireless Power Transfer (WPT) Systems for Powering Implanted Biomedical Devices (IBDs) using Rogers (RO4003C) material. The rectifier is employed as an actual DC-supply voltage for IBDs. The wide-band operation is achieved through a two-branch design, requiring careful design of both branches to ensure seamless performance across the frequency range of 0.2 to 0.9 GHz. The first-branch rectifier operates at 433 MHz, while the second-branch rectifier works at 675 MHz. The design is considered as two cascade voltage doublers (VD), where the output of the first VD stage serves as the ground or virtual DC-reference for the subsequent stage. The closed-coupled transmission lines (CTLs) are used at the design input for efficiency and voltageboosting, especially to increase the efficiency in the middle-frequency band by 10 %. The rectifier achieves a maximum simulated efficiency of 72 % and a DC-voltage (<span><math><msub><mi>V</mi><mrow><mi>d</mi><mi>c</mi></mrow></msub></math></span>) equals 3.5 V at input power (<span><math><msub><mi>P</mi><mrow><mi>i</mi><mi>n</mi></mrow></msub></math></span>) equal 4 dBm. It demonstrates a measured efficiency of 60 % and a measured <span><math><mrow><msub><mi>V</mi><mrow><mi>d</mi><mi>c</mi></mrow></msub><mo>=</mo><mn>3.5</mn><mi>V</mi></mrow></math></span> at <span><math><mrow><msub><mi>P</mi><mrow><mi>i</mi><mi>n</mi></mrow></msub><mo>=</mo><mn>4</mn><mtext>dBm</mtext></mrow></math></span>. This design demonstrates efficiency <span><math><mrow><mo>&gt;</mo><mn>40</mn><mrow><mspace></mspace><mo>%</mo><mspace></mspace></mrow></mrow></math></span>with a large <span><math><mrow><mi>R</mi><mo>=</mo><mn>18</mn><mi>d</mi><mi>B</mi></mrow></math></span>, and a compact area of <span><math><mrow><mn>4.8</mn><mi>c</mi><msup><mrow><mi>m</mi></mrow><mn>2</mn></msup></mrow></math></span> .</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110708"},"PeriodicalIF":4.9,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating perceptual quality analysis and caption-based features for robust deepfake video detection","authors":"Hessen Bougueffa Eutamene ,&nbsp;Wassim Hamidouche ,&nbsp;Mamadou Keita ,&nbsp;Abdelmalik Taleb-Ahmed ,&nbsp;Abdenour Hadid","doi":"10.1016/j.compeleceng.2025.110699","DOIUrl":"10.1016/j.compeleceng.2025.110699","url":null,"abstract":"<div><div>Inspired by the benefits of both linguistic representations as well as visual quality analysis, we propose a <strong>multimodal framework for deepfake detection with reliable accuracy</strong>. Our method acquires low-level perceptual features from frames of video, including contrast, brightness, and sharpness, and computes artifact scores for artificial anomaly detection. In parallel, our method produces descriptive frame-level captions, summed up for generating video-level summaries for capturing contextual coherence. For training the model, our method leverages the <strong>FaceForensics++</strong> dataset, comprising several techniques of deepFake manipulation, including DeepFake, Face2Face, FaceSwap, as well as NeuralTextures. Metadata comprised of quality measurements, the artifact scores, as well as textual captions, is tokenized, processed by a <strong>DeepSeek V2 Lite model</strong>, fine-tuned by the <strong>Low-Rank Adaptation (LoRA)</strong> procedure, as the backbone for classification. We test the proposed model on both intra-dataset as well as cross-dataset settings, demonstrating good accuracy as well as impressive generalization ability. Our method achieves a <strong>96.51%</strong> accuracy for classifying, by integrating low-level perceptual as well as high-level semantic reason, effectively marking manipulated content as well as upholding the integrity of the medium in a world of increasingly evolving generative methods. The employed source code as well as dataset preprocessing scripts for our work are made available for public access on: <span><span>https://github.com/HessenUPHF/perceptual-caption-detection</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110699"},"PeriodicalIF":4.9,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A shallow hybrid model with dynamic Bayesian optimisation for wind speed prediction on memory-constrained devices","authors":"Laeeq Aslam ,&nbsp;Runmin Zou ,&nbsp;Yaohui Huang ,&nbsp;Fatima Yaqoob ,&nbsp;Sharjeel Abid Butt ,&nbsp;Qian Zhou","doi":"10.1016/j.compeleceng.2025.110700","DOIUrl":"10.1016/j.compeleceng.2025.110700","url":null,"abstract":"<div><div>Accurate wind speed prediction (WSP) supports energy management, but many deep learning models are too computationally intensive for embedded systems. This work presents a shallow hybrid model optimised with a dynamic tree-structured Parzen estimator. The model employs fast independent component analysis and an adaptive least absolute shrinkage and selection operator for feature selection, alongside a concurrent recurrent temporal module with a random vector functional link layer. The output is then refined through an attention layer to capture temporal dependencies. Across three sites, the model reduces mean squared error by up to 5.9% and mean absolute error by 5.03%, while increasing the coefficient of determination by 1.14 percentage points compared with the strongest baseline, alongside a 38.04% reduction in parameters. This design balances accuracy with resource efficiency and is well-suited to memory-constrained applications.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110700"},"PeriodicalIF":4.9,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancements in acne detection and classification: A bibliometric analysis","authors":"Krishna Veni Paluri ,&nbsp;Garima Nain ,&nbsp;Ashish Gupta","doi":"10.1016/j.compeleceng.2025.110693","DOIUrl":"10.1016/j.compeleceng.2025.110693","url":null,"abstract":"<div><div>Acne is a chronic inflammatory dermatological condition that presents notable diagnostic issues related to its intricate phenotypic features, especially in its initial phases. Leveraging Artificial Intelligence (AI) in dermatology has demonstrated the capability to enhance diagnostic accuracy, improve clinical results, and provide scalable treatments. This bibliometric analysis investigates the utilization of Machine Learning (ML) and Deep Learning (DL) techniques in acne detection, emphasizing progress in the analysis of medical images and equipment for diagnosis. Utilizing VOSviewer and Biblioshiny, 549 publications from the Scopus database (2016–2025) were examined to identify trends in publications and citations, co-authorship networks, keyword co-occurrence, and bibliographic coupling. The analysis highlighted thematic clusters (nine clusters), emphasizing essential research domains and limitations. The results emphasize the significance of innovation and collaboration among researchers, healthcare professionals, and policymakers for enhancing patient outcomes. It also provides the taxonomy of AI methods in Acne detection and classification. This study explores the current status of AI applications in acne detection and classification, intending to drive future research and advancements in technology, thereby enhancing the effectiveness and accessibility of dermatological care.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"128 ","pages":"Article 110693"},"PeriodicalIF":4.9,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}