IEEE Access最新文献

{"title":"MICAAL: A Domain-Specific Language for Microservices in Ambient Assisted Living","authors":"Wilson Valdez Solis;Priscila Cedillo;Attila Kertesz","doi":"10.1109/ACCESS.2025.3555831","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3555831","url":null,"abstract":"The rise of the Internet of Things (IoT) has driven innovation across various sectors, one of its benefiting areas is healthcare. Within this domain, Ambient Assisted Living (AAL) aims to improve the wellbeing and independence of vulnerable populations, particularly seniors, by using IoT technologies. As AAL environments become increasingly complex and diverse, traditional monolithic architectures struggle to offer the flexibility and scalability required for sustainable solutions. In response, paradigms such as microservices architectures emerge as a powerful approach to overcome these challenges. This paper presents MICAAL, a Domain-Specific Language (DSL), to model microservices-oriented architectures for AAL environments. MICAAL empowers users to design modular, scalable systems by enabling the easy selection and deployment of microservices ensuring that IoT devices in AAL environments are efficiently supported and adaptable to changing needs. We demonstrate the practical usability of MICAAL through a case study of a real-world AAL system. We also present an empirical evaluation, using the Technology Acceptance Model (TAM), assessing user satisfaction and intention to adopt MICAAL for modeling microservice-based architectures for AAL. The results confirm that MICAAL simplifies the modeling process and promotes efficient, future-proof AAL systems.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"56255-56272"},"PeriodicalIF":3.4,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945329","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Single-Camera-Based 3D Drone Trajectory Reconstruction for Surveillance Systems","authors":"Seo-Bin Hwang;Yeong-Jun Cho","doi":"10.1109/ACCESS.2025.3555321","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3555321","url":null,"abstract":"Drones have been utilized in various fields, but the number of drones being used illegally and for illegal and hazardous purposes has recently increased. These misuses have given rise to the development of an anti-drone system, C-UAS, comprising two key steps: engagement and detection. This detection step plays a crucial role in reconstructing the drone’s trajectory for the subsequent engagement step. Therefore, we focus on a trajectory reconstruction approach that utilizes external data, such as CCTV view images, instead of internal drone data. While numerous methods have been explored for estimating 3D drone trajectory using multiple sensors, they are often unsuitable for surveillance systems. In this study, we use a calibrated single camera suitable for surveillance systems, leveraging the relationship between 2D and 3D spaces. We use a drone tracker to automatically track 2D images. The tracked images are used to estimate the 2D rotation of the drone in the image through principal component analysis (PCA). By combining the estimated 2D drone positions with the actual length, we geometrically infer the 3D drone trajectories. We also develop synthetic 2D and 3D drone datasets to address the lack of public drone datasets. Additionally, a real-world 3D dataset is generated and made available for public use. The experimental results demonstrate that the proposed method can accurately reconstruct drone trajectories in 3D space. With a MAE of 5.66 and RMSE of 7.93 showing low error rates, these findings validate the practical value and potential of our framework as a single-camera-based surveillance system.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"56413-56427"},"PeriodicalIF":3.4,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10942597","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Competency-Based Hybrid Learning: A Modern Approach to Teaching Programming and Digital Technologies Subjects","authors":"Erik Kučera;Oto Haffner","doi":"10.1109/ACCESS.2025.3555333","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3555333","url":null,"abstract":"The COVID-19 pandemic prompted the global education sector to experiment with various forms of online learning as institutions rapidly transitioned to remote formats. This article presents a comprehensive overview of a competency-based hybrid learning methodology, developed and implemented during the pandemic. In addition to detailing the methodology itself, the article also shares the experiences of educators at the authors’ institution, who observed significant improvements in educational outcomes, surpassing even pre-pandemic standards. The methodology highlights the limitations of directly replicating traditional in-person instruction in an online format using existing materials and approaches. Instead, it advocates for carefully designed adaptations tailored to the digital environment, leveraging asynchronous components, interactive tools, and newly created e-learning resources to optimize effectiveness. This approach also requires increased interaction between educators and students beyond scheduled classes, ensuring timely support and guidance. Although this methodology may not suit all course types, it has proven particularly effective in advanced information and communication technologies (ICT) and digital technology subjects, such as programming, artificial intelligence, digital or electronic marketing, video editing, 3D engine work, etc. The positive student feedback further underscores the potential of this model to enhance educational quality and outcomes in these domains.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"54892-54919"},"PeriodicalIF":3.4,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10943157","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Discrete Predefined-Time Terminal Sliding Mode Controller for Current Control of PMSM Drives","authors":"Haibo Xue;Xinghua Liu","doi":"10.1109/ACCESS.2025.3555483","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3555483","url":null,"abstract":"This paper focuses on the discrete sliding mode control (SMC) method adopted in the permanent magnet synchronous motor (PMSM) current loop control system. Based on the predefined-time terminal sliding mode (PTSM) method, a discrete PTSM and its sufficient conditions for convergence are proposed. Considering that the PMSM current loop control system can be rewritten as a typical discrete first-order multiple-input multiple-output (MIMO) system in practical applications, a discrete PTSM law is further proposed. The effectiveness of the proposed discrete PTSM controller is verified through numerical simulations and experiments on a PMSM drive control test bench. Experiments are conducted to compare different current loop control schemes, which fully demonstrates that the discrete PTSM method proposed in this paper can achieve better dynamic performance of discrete state variables during the convergence sliding process.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"56273-56282"},"PeriodicalIF":3.4,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10943164","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Improving Investment Forecasting: A Comparative Analysis of Machine Learning Models as Key GDP Indicators","authors":"Cheng-Hong Yang;Tshimologo Molefyane;Borcy Lee;Ting-Jen Hsueh;Yu-da Lin","doi":"10.1109/ACCESS.2025.3554641","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3554641","url":null,"abstract":"Investment is a crucial driver of economic growth and a fundamental component of Gross Domestic Product (GDP). Accurate investment forecasting is essential for informed policymaking and economic planning. However, traditional econometric models often struggle to capture economic data’s intricate and non-linear dynamics, indicating a need for more robust and adaptable approaches. This study addresses this need by employing machine learning techniques to enhance investment forecasting accuracy, particularly the Gated Recurrent Unit (GRU) model. Historical investment data from fifteen leading GDP countries (1990–2020) was analyzed using seven models: GRU, ARIMA, ETS, SVR, XGBoost, CNN, and LSTM. Key performance metrics—Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE)—demonstrated that the GRU model achieved superior accuracy. The GRU model consistently outperformed other models across all performance metrics. It achieved the lowest MAE, RMSE, and MAPE values, highlighting its effectiveness in capturing complex temporal dependencies. Compared to traditional econometric models, GRU delivered significantly more accurate forecasts, emphasizing its potential for improving investment predictions. The results emphasize the potential of advanced machine learning models in capturing complex temporal dependencies, leading to more reliable investment predictions. This study addresses a significant gap in economic forecasting by highlighting the advantages of GRU and its implications for policymaking and future research.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"54517-54533"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10938555","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Exploring Beyond Logits: Hierarchical Dynamic Labeling Based on Embeddings for Semi-Supervised Classification","authors":"Jiayi Chen;Yanbiao Ma;Wei Dai;Xiaohua Chen;Shuo Li","doi":"10.1109/ACCESS.2025.3554583","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3554583","url":null,"abstract":"In semi-supervised learning, methods that rely on confidence learning to generate pseudo-labels have been widely proposed. However, increasing research finds that when faced with noisy and biased data, the model’s representation network is more reliable than the classification network. Additionally, label generation methods based on model predictions often show poor adaptability across different datasets, necessitating customization of the classification network. Therefore, we propose a Hierarchical Dynamic Labeling (HDL) algorithm that does not depend on model predictions and utilizes image embeddings to generate sample labels. We also introduce an adaptive method for selecting hyperparameters in HDL, enhancing its versatility. Moreover, HDL can be combined with general image encoders (e.g., CLIP) to serve as a fundamental data processing module. We extract embeddings from datasets with class-balanced and long-tailed distributions using pre-trained semi-supervised models. Subsequently, samples are re-labeled using HDL, and the re-labeled samples are used to further train the semi-supervised models. Experiments demonstrate improved model performance, validating the motivation that representation networks are more reliable than classifiers or predictors. Our approach has the potential to change the paradigm of pseudo-label generation in semi-supervised learning.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"54611-54621"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10938612","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Modeling the Error of Caliper Measurements in Animal Experiments","authors":"Melánia Puskás;Dániel András Drexler","doi":"10.1109/ACCESS.2025.3555148","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3555148","url":null,"abstract":"Cancer prevention and treatment is one of the most significant public health challenges of the 21st century. Cancer is a serious health problem, and it is the second leading cause of death following cardiovascular diseases. This requires a reliable virtual patient model, which is usually created based on animal studies that precede human studies. Preclinical drug testing often involves mouse experiments, where tumors are implanted under the skin. Up until now, the most widespread tumor measurement method is caliper measurement, which involves a large measurement error, especially if the tumor is small. We present a noise model that can be used to model the measurement noise in animal experiments where tumor size is measured with calipers. Accurate in silico measurement is essential, as animal studies are costly, time-consuming, and strictly regulated. By incorporating a noise model, in silico experiments can better reflect real-world measurement uncertainties, improving experimental reproducibility and the reliability of virtual patient modeling. In order to model the noise, we use data from preclinical experiments measured using MRI and digital calipers, and we use a nonlinear transformation to whiten the noise. Finally, based on the Anderson-Darling test, we find the distributions that fit the noise best. We show that virtually generated measurements based on the noise model produce similar results to the original measurement noise, thus the noise model can be used to create virtual patients and model realistic experimental setups for in silico experiments.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"54836-54852"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10942315","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Unsupervised Video Anomaly Detection Using Video Vision Transformer and Adversarial Training","authors":"Shimpei Kobayashi;Akiyoshi Hizukuri;Ryohei Nakayama","doi":"10.1109/ACCESS.2025.3554813","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3554813","url":null,"abstract":"Surveillance cameras have been recently introduced in various locations to maintain public safety. However, it is tedious for security personnel to continue observing videos obtained by surveillance cameras because abnormal events rarely occur. Consequently, this study aims to develop an unsupervised anomaly detection method for surveillance videos based on the prediction of future frame and optical flow magnitude using adversarial networks. Our database consists of two public datasets for anomaly detection: UCSD Pedestrian 2 (UCSD Ped2) and ShanghaiTech Campus. The proposed network consists of a generator and a U-Net-based discriminator. The generator has an encoder based on a video vision transformer and two decoders that independently predict the future frame and optical flow magnitude using feature maps extracted from the encoder at different resolutions. The U-Net-based discriminator identifies at the pixel level whether the image is a “real” or “false” frame predicted by the generator. The generator and U-Net-based discriminator in the proposed network are trained through adversarial training. Finally, anomaly detection is performed based on element-wise subtraction between the predicted future frame and the optical flow magnitude and their corresponding ground truth frames. The areas under the receiver operating characteristic curve (AUC) with the proposed network were 97.5% for UCSD Ped2 and 85.8% for ShanghaiTech Campus. The AUC for the proposed network was also greater than those of well-known anomaly detection methods.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"56169-56179"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10942323","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"EFFResNet-ViT: A Fusion-Based Convolutional and Vision Transformer Model for Explainable Medical Image Classification","authors":"Tahir Hussain;Hayaru Shouno;Abid Hussain;Dostdar Hussain;Muhammad Ismail;Tatheer Hussain Mir;Fang Rong Hsu;Taukir Alam;Shabnur Anonna Akhy","doi":"10.1109/ACCESS.2025.3554184","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3554184","url":null,"abstract":"The rapid advancement of medical imaging technologies requires the development of advanced, automated, and interpretable diagnostic tools for clinical decision-making. Although convolutional neural networks (CNNs) have shown significant promise in medical image analysis, they have limitations in capturing the global context and lack interpretability, thereby hindering their clinical adoption. This study presents EFFResNet-ViT, a novel hybrid deep learning (DL) model designed to address these challenges by combining EfficientNet-B0 and ResNet-50 CNN backbones with a vision transformer (ViT) module. The proposed architecture employs a feature fusion strategy to integrate the local feature extraction strengths of CNNs with the global dependency modeling capabilities of transformers. The extracted features are further refined through a post-transformer CNN and a global average pooling layer to enhance the classification performance. To improve interpretability, EFFResNet-ViT incorporates Grad-CAM visualization techniques to highlight regions contributing to classification decisions and employs t-distributed stochastic neighbor embedding for feature space analysis, providing insights into class separability. The proposed model was evaluated on two benchmark datasets: brain tumor (BT) CE-MRI for BT classification and a retinal image dataset for ophthalmological diagnosis. EFFResNet-ViT achieved state-of-the-art performance, with accuracies of 99.31% and 92.54% on the BT CE-MRI and retinal datasets, respectively. Comparative analyses demonstrate the superior classification performance and interpretability of EFFResNet-ViT over existing ViT and CNN-based hybrid models. The explainable design of EFFResNet-ViT addresses the critical need for transparency in artificial intelligence-driven medical diagnostics, facilitating its potential integration into clinical workflows to improve decision-making and patient outcomes.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"54040-54068"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10938132","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"End-to-End Deployment of the Educational AI Hub for Personalized Learning and Engagement: A Case Study on Environmental Science Education","authors":"Ramteja Sajja;Yusuf Sermet;Ibrahim Demir","doi":"10.1109/ACCESS.2025.3554222","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3554222","url":null,"abstract":"This study introduces an end-to-end framework for deploying conversational AI-enabled educational assistants, focusing on personalized support for students across diverse subject areas, including Business, Culture, Environmental Sciences, History, Politics, and Science, as outlined in our evaluation framework. The system leverages advanced conversational AI technologies to provide targeted, course-specific learning experiences by facilitating access to complex data and integrating seamlessly with Learning Management Systems (LMS) like Canvas. Key metrics—information retrieval accuracy, question-answering accuracy, and hallucination accuracy—were selected to rigorously evaluate the system’s ability to retrieve relevant contexts, generate accurate responses, and identify unanswerable questions. Additionally, the Educational AI Hub agents utilize innovative document parsing methods, such as the Nougat technique, to interpret content accurately, enabling adaptable academic support tailored to individual learning needs and extending to quantitative fields through code execution capabilities. This study also emphasizes the importance of accessibility, inclusivity, and user privacy. The results showcase the potential for enhanced engagement and improved understanding of environmental concepts and software tools, demonstrating the significant impact of conversational AI in educational settings, especially in disciplines involving complex data interactions. A case study, presented at the 12th International Congress on Environmental Modelling and Software, illustrates the Educational AI Hub’s effectiveness in enhancing student engagement and delivering personalized learning experiences in environmental sciences education.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"55169-55186"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10938135","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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}