计算机科学最新文献

{"title":"Circular Polarized Reconfigurable Design of Elliptical Microstrip Antenna for GPS L-Band Applications","authors":"Amit A. Deshmukh,&nbsp;Heet Mistry,&nbsp;Venkata A. P. Chavali","doi":"10.1002/dac.70232","DOIUrl":"https://doi.org/10.1002/dac.70232","url":null,"abstract":"<div>\u0000 \u0000 <p>Frequency reconfigurable antennas are required in wireless applications, as they cover a wide spectrum addressing multiple bands, but with a smaller antenna size. The proposed study presents open circuit stubs loaded elliptical shape microstrip antenna design to provide tunable circular polarized response that covers various GPS L-bands. A reconfigurable approach is used that selects the interconnection of stubs of increasing lengths, achieved through the suitable excitation of PIN diodes. Through this, orthogonal modes on the elliptical patch are tuned in their frequencies that provide a tuning frequency range from 1575 MHz (L1) to 1176 MHz (L5), thus addressing the complete global positioning system (GPS) spectrum. The antenna offers a broadside pattern in each band with axial ratio bandwidth greater than 2.5% and peak gain of more than 7 dBi, thereby satisfying the requirements of the GPS L-band system. An experimental validation is carried out for the proposed designs that show close agreement against the simulated results.</p>\u0000 </div>","PeriodicalId":13946,"journal":{"name":"International Journal of Communication Systems","volume":"38 14","pages":""},"PeriodicalIF":1.8,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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":"Adaptive confidence-driven learning and cross-modal hard sample mining for unsupervised visible-infrared person re-identification","authors":"Yifeng Zhang ,&nbsp;Canlong Zhang ,&nbsp;Haifei Ma ,&nbsp;Zhixin Li ,&nbsp;Zhiwen Wang ,&nbsp;Chunrong Wei","doi":"10.1016/j.ipm.2025.104346","DOIUrl":"10.1016/j.ipm.2025.104346","url":null,"abstract":"<div><div>This research addresses the critical challenges in Cross-modal Visible-Infrared Person Re-ID (VI-ReID), including significant modal differences, lack of cross-modal correspondence, and pseudo-label noise accumulation. To mitigate these issues, we propose an innovative framework integrating an adaptive multidimensional enhanced clustering method and a confidence-driven dynamic label correction mechanism. Specifically, we design a dynamic clustering framework leveraging neighborhood consistency and intra-class distribution entropy to autonomously model data distributions. A confidence-driven dynamic label correction mechanism is introduced, employing multi-prototype similarity probability models to filter pseudo-label noise effectively. Moreover, a cross-modal feature alignment strategy based on optimal transport theory addresses many-to-many feature matching between visible and infrared modalities. Additionally, a Hard Sample Aware Contrastive Learning (HCL) strategy is implemented to enhance feature learning in complex data distributions through dynamic feature storage. Extensive experiments conducted on SYSU-MM01 and RegDB datasets, comprising 29,533 and 4120 image pairs, respectively, demonstrate the framework’s effectiveness. The proposed method achieves a 3.9% mAP improvement on average compared to state-of-the-art methods, highlighting its advantages in cross-modal feature alignment and pseudo-label optimization.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 1","pages":"Article 104346"},"PeriodicalIF":6.9,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144841886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fault diagnosis of air handling units based on an MCNN-Transformer ensemble learning","authors":"Yin Xia,&nbsp;Danhong Zhang,&nbsp;Chenyu Liu,&nbsp;Zhiqiang Cao,&nbsp;Yixin Su,&nbsp;Yuhang Chen","doi":"10.1016/j.jprocont.2025.103526","DOIUrl":"10.1016/j.jprocont.2025.103526","url":null,"abstract":"<div><div>The Air Handling Units (AHU) in Heating Ventilation and Air Conditioning (HVAC) systems regulates air temperature and humidity to ensure indoor air quality and thermal comfort. Fault diagnosis of AHU is critical for reducing energy consumption and maintaining system performance. However, data noise and missing values introduce considerable uncertainty into AHU fault diagnosis, while most existing methods do not utilize time-series models and thus neglect the extraction of temporal features and the modeling of long-range dependencies. This limitation hinders the effective capture of fault evolution and long-term correlations, making it difficult to meet dynamic real-time requirements under complex operating conditions. To address these challenges, this paper proposes an ensemble learning framework that integrates Dempster–Shafer (DS) theory with a Multi-Channel Convolutional Neural Network and Transformer (MCNN-Transformer) model, aiming to enhance generalization and improve diagnostic performance. The DS theory combines the strengths of Random Forest, Pearson Correlation, and Mutual Information, effectively mitigating uncertainty and noise in fault feature data by fusing multi-source information. The MCNN-Transformer integrates multi-scale convolutional layers with a self-attention mechanism, enabling effective extraction of features across multiple temporal scales and modeling of long-range dependencies. Experimental results show that the proposed MCNN-Transformer framework achieves high efficiency and strong generalization capability, reaching a fault diagnosis accuracy of 99.2%, a precision of 0.992, a recall of 0.992, and an F1 score of 0.991, significantly outperforming traditional models. Moreover, the improved stability of the model’s accuracy curve further demonstrates its robustness.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"154 ","pages":"Article 103526"},"PeriodicalIF":3.9,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 Novel Multimodal Medical Image Fusion Method Based on Detail Enhancement and Dual-Branch Feature Fusion","authors":"Kun Zhang,&nbsp;Hui Yuan,&nbsp;Zhongwei Zhang,&nbsp;PengPeng Sun","doi":"10.1002/ima.70181","DOIUrl":"https://doi.org/10.1002/ima.70181","url":null,"abstract":"<div>\u0000 \u0000 <p>Multimodal medical image fusion integrates effective information from different modal images and integrates salient and complementary features, which can more comprehensively describe the condition of lesions and make medical diagnosis results more reliable. This paper proposes a multimodal medical image fusion method based on image detail enhancement and dual-branch feature fusion (DEDF). First, the source images are preprocessed by guided filtering to enhance important details and improve the fusion and visualization effects. Then, local extreme maps are used as guides to smooth the source images. Finally, a DEDF mechanism based on guided filtering and bilateral filtering is established to obtain multiscale bright and dark feature maps, as well as base images of different modalities, which are fused to obtain a more comprehensive medical image and improve the accuracy of medical diagnosis results. Extensive experiments, compared qualitatively and quantitatively with various state-of-the-art medical image fusion methods, validate the superior fusion performance and effectiveness of the proposed method.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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":"An Integrated Social Robot and Virtual Assistant Solution to Support Medical Management for Older Adults","authors":"Matheus Ancelmo Bonfim Pita,&nbsp;Marcelo Fantinato,&nbsp;Patrick C. K. Hung","doi":"10.1111/exsy.70112","DOIUrl":"https://doi.org/10.1111/exsy.70112","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>The global aging population leads to increased demand for professional caregivers and innovative assistive technologies. Traditional aids such as canes and hearing devices have long supported older adults, but emerging solutions involving robotics and AI open new opportunities for enhanced care and independence.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study aimed to design and evaluate an assistive solution that integrates a social robot and a virtual assistant to support older adults in managing medical treatments and daily schedules.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>An assistive system was developed combining a social robot and a virtual assistant. Its potential was assessed through an exploratory evaluation involving seven older adults who interacted with the solution in simulated care and schedule management scenarios. Data were collected through structured interviews to capture participants' perceptions and experiences.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The developed solution supported effective interaction between users and the technologies, despite minor usability challenges during initial use. Participants were generally able to complete tasks such as medication reminders, appointment management, and basic conversational interactions, although some required occasional assistance or clarification.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Evaluation</h3>\u0000 \u0000 <p>The participants expressed positive feedback regarding usability and perceived usefulness. The combined use of social robots and virtual assistants was considered intuitive and supportive, especially in reducing cognitive load and fostering adherence to treatment routines.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>The integrated assistive solution presents a promising approach to supporting older adults' independence and well-being. By combining social presence with functional assistance, it contributes to bridging the gap between human-centered care and technological innovation.</p>\u0000 </section>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 9","pages":""},"PeriodicalIF":2.3,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/exsy.70112","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"BATU: A Workflow for Multi-Network Ensemble Learning in Cross-Dataset Generalization of Skin Lesion Analysis","authors":"Ömer Faruk Söylemez","doi":"10.1002/ima.70183","DOIUrl":"https://doi.org/10.1002/ima.70183","url":null,"abstract":"<div>\u0000 \u0000 <p>The development of computer vision systems for dermatological diagnosis is often hindered by dataset heterogeneity, including differences in image quality, labeling strategies, and patient demographics. In this study, we examine how such heterogeneity affects the generalization ability of computer vision models across three public dermatology image datasets. We trained five different deep learning models on each dataset separately and evaluated their performance in both intra-dataset and cross-dataset settings. To further investigate robustness, we conducted multi-source domain generalization experiments by training models on combinations of two datasets and testing on the third unseen dataset. We observed a significant drop in performance during cross-dataset evaluations. To address this, we applied various ensemble learning methods by combining the predictions from the individual models. Our results demonstrate that ensemble approaches consistently outperform individual models, achieving accuracy improvements exceeding 4% in many cases. These findings highlight the potential of ensemble learning to address challenges related to dataset variability in dermatological image analysis.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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":"Machine learning-optimized stochastic Voronoi lattices for enhanced mechanical performance","authors":"Michael Thompson ,&nbsp;Hamidreza Yazdani Sarvestani ,&nbsp;Ahmad Sohrabi-Kashani ,&nbsp;Elham Kiyani ,&nbsp;Enzo Filippi ,&nbsp;Derek Aranguren van Egmond ,&nbsp;Meysam Rahmat ,&nbsp;Behnam Ashrafi ,&nbsp;Mikko Karttunen","doi":"10.1016/j.engappai.2025.111937","DOIUrl":"10.1016/j.engappai.2025.111937","url":null,"abstract":"<div><div>Lattice structures, traditionally composed of periodic networks of interconnected struts, offer an excellent balance of high strength and low density. However, their periodicity limits adaptability to complex and unpredictable loading conditions. Stochastic Voronoi lattices, characterized by their irregular, non-periodic geometry, provide a promising alternative with enhanced energy absorption and mechanical robustness. In this study, we present a machine learning (ML)-driven framework integrating finite element analysis (FEA), a multilayer perceptron (MLP) neural network, and three-dimensional (3D) printing to optimize Voronoi lattice structures for targeted mechanical properties. To systematically control structural disorder, we introduce Relaxation Iteration (RI), an ordering parameter inspired by Lloyd’s algorithm. Based on RI, we show that there is a range of <span><math><mrow><mi>RI</mi><mo>≈</mo><mn>1500</mn><mo>−</mo><mn>2000</mn></mrow></math></span> which gives enhanced mechanical performance for Voronoi lattices. The ML-FEA optimized Voronoi lattices demonstrate double the stiffness and four times the toughness compared to conventional periodic lattices. These findings underscore the potential of ML-driven design strategies in developing tailored architected materials for applications requiring high energy absorption and structural integrity, including aerospace, automotive crash protection, and biomedical implants.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"160 ","pages":"Article 111937"},"PeriodicalIF":8.0,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144840779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Is generative AI reshaping academic practices worldwide? A survey of adoption, benefits, and concerns","authors":"Ehsan Mohammadi ,&nbsp;Mike Thelwall ,&nbsp;Yizhou Cai ,&nbsp;Taylor Collier ,&nbsp;Iman Tahamtan ,&nbsp;Azar Eftekhar","doi":"10.1016/j.ipm.2025.104350","DOIUrl":"10.1016/j.ipm.2025.104350","url":null,"abstract":"<div><div>Although generative AI is transforming academic research and education, little is known about the role, gender, international, and disciplinary variations in uptake and use. This 20-country survey of publishing academics shows the widespread awareness and adoption of generative AI tools in academia, but with substantial international and disciplinary differences, and some role and gender differences. In particular, females were 10 % less likely to use Gen AI frequently (daily or weekly) for research, which may exacerbate gender inequalities. Perhaps surprisingly, the highest adoption rates occurred in some non-Western nations, possibly because of a greater need for translation services. The highest awareness is in the social sciences, perhaps because of the greater need for text analysis. Across all groups, these tools were mainly used for academic writing rather than data analysis and support for critical thinking. Despite this, personalized instruction and problem-solving are among generative AI's most generally claimed benefits. However, participants in all groups were skeptical about the creativity, accuracy, and consistency of AI-generated content in academic contexts. The most significant concerns about using generative AI in academia were inaccuracy, plagiarism, discouraging critical thinking, a lack of transparency and explainability, intellectual property rights violations, and data privacy risks. For policymakers, the findings point to fields and countries that may need action to prevent falling behind, as well as the ongoing need to investigate and monitor the impacts of generative AI on research practices.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 1","pages":"Article 104350"},"PeriodicalIF":6.9,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144841887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced Computational Complexity in Continuous-Depth Models: Neural Ordinary Differential Equations With Trainable Numerical Schemes.","authors":"Said Ouala,Laurent Debreu,Bertrand Chapron,Fabrice Collard,Lucile Gaultier,Ronan Fablet","doi":"10.1109/tpami.2025.3599629","DOIUrl":"https://doi.org/10.1109/tpami.2025.3599629","url":null,"abstract":"Neural Ordinary Differential Equations (NODEs) serve as continuous-time analogs of residual networks. They provide a system-theoretic perspective on neural network architecture design and offer natural solutions for time series modeling, forecasting, and applications where invertible neural networks are essential. However, these models suffer from slow performance due to heavy numerical solver overhead. For instance, a popular solution for training and inference of NODEs consists in using adaptive step size solvers such as the popular Dormand-Prince 5(4) (DOPRI). These solvers dynamically adjust the Number of Function Evaluations (NFE) as the equation fits the training data and becomes more complex. However, this comes at the cost of an increased number of function evaluations, which reduces computational efficiency. In this work, we propose a novel approach: making the parameters of the numerical integration scheme trainable. By doing so, the numerical scheme dynamically adapts to the dynamics of the NODE, resulting in a model that operates with a fixed NFE. We compare the proposed trainable solvers with state-of-the-art approaches, including DOPRI, for different benchmarks, including classification, density estimation, and dynamical system modeling. Overall, we report a state-of-the-art performance for all benchmarks in terms of accuracy metrics, while enhancing the computational efficiency through trainable fixed-step-size solvers. This work opens up new possibilities for practical and efficient modeling applications with NODEs.","PeriodicalId":13426,"journal":{"name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","volume":"80 1","pages":""},"PeriodicalIF":23.6,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144857764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Density peak clustering based on nearest neighbors","authors":"Houshen Lin,&nbsp;Jian Hou,&nbsp;Huaqiang Yuan","doi":"10.1016/j.engappai.2025.111981","DOIUrl":"10.1016/j.engappai.2025.111981","url":null,"abstract":"<div><div>As a promising clustering approach, the density peak clustering (DPC) algorithm is sensitive to density kernels and involved parameters, and its non-central data allocation method is likely to result in error propagation. In order to solve these two problems in dealing with real-world data, in this paper we present a new density peak clustering algorithm by making full use of the concept of <span><math><mi>k</mi></math></span>-nearest neighbors. In density estimation, we present a Gaussian kernel constrained by <span><math><mi>k</mi></math></span>-nearest neighbors and determine the cutoff distance based on local data distribution adaptively. In non-central data allocation, we propose a method to avoid error propagation by using the <span><math><mi>k</mi></math></span>-nearest neighbors of allocated data and unallocated data alternately. Starting from the cluster centers, we search the <span><math><mi>k</mi></math></span>-nearest neighbors of allocated data points, and include the qualified nearest neighbors into clusters. For each unallocated data point, we determine its belonging based on the allocated data points in its <span><math><mi>k</mi></math></span>-nearest neighbors. After that, we do a re-allocation of some allocated data points based on the allocated data points in their <span><math><mi>k</mi></math></span>-nearest neighbors, to improve the clustering accuracy further. In experiments on 50 synthetic and real datasets, our algorithm generates better results than 14 DPC-based and 6 non-DPC based algorithms. By making full use of <span><math><mi>k</mi></math></span>-nearest neighbors in all the major steps and exploiting the power of Gaussian distribution in real-world data processing, our algorithm is shown to be promising in dealing with real-world data clustering.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"160 ","pages":"Article 111981"},"PeriodicalIF":8.0,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144840778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}