Pub Date : 2025-01-22DOI: 10.1109/ACCESS.2024.3525276
Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.
提供社会信息，可能包括新闻，评论或技术笔记，从业者和研究人员应该感兴趣。
{"title":"IEEE Access™ Editorial Board","authors":"","doi":"10.1109/ACCESS.2024.3525276","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3525276","url":null,"abstract":"Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"i-x"},"PeriodicalIF":3.4,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10849622","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993421","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}
Pub Date : 2025-01-17DOI: 10.1109/ACCESS.2024.3518212
M. Poongodi;Mohit Malviya;Mounir Hamdi;Hafiz Tayyab Rauf;Seifedine Kadry;Orawit Thinnukool
Presents corrections to the paper, (Corrections to “The Recent Technologies to Curb the Second-Wave of COVID-19 Pandemic”).
对论文进行更正（对“遏制第二波COVID-19大流行的最新技术”的更正）。
{"title":"Corrections to “The Recent Technologies to Curb the Second-Wave of COVID-19 Pandemic”","authors":"M. Poongodi;Mohit Malviya;Mounir Hamdi;Hafiz Tayyab Rauf;Seifedine Kadry;Orawit Thinnukool","doi":"10.1109/ACCESS.2024.3518212","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3518212","url":null,"abstract":"Presents corrections to the paper, (Corrections to “The Recent Technologies to Curb the Second-Wave of COVID-19 Pandemic”).","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"5436-5436"},"PeriodicalIF":3.4,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10839147","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993009","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}
Pub Date : 2025-01-17DOI: 10.1109/ACCESS.2025.3527322
Fahad Tanveer;Muhammad Bilal Kadri
Presents corrections to the paper, Corrections to “Decentralized Asynchronous Formation Planning of Multirotor Aerial Vehicles in Dynamic Environments Using Flexible Formation Graphs and Tight Trajectory Hulls”.
对“动态环境下多旋翼飞行器基于柔性编队图和紧密弹道壳体的分散异步编队规划”论文进行了修正。
{"title":"Corrections to “Decentralized Asynchronous Formation Planning of Multirotor Aerial Vehicles in Dynamic Environments Using Flexible Formation Graphs and Tight Trajectory Hulls”","authors":"Fahad Tanveer;Muhammad Bilal Kadri","doi":"10.1109/ACCESS.2025.3527322","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527322","url":null,"abstract":"Presents corrections to the paper, Corrections to “Decentralized Asynchronous Formation Planning of Multirotor Aerial Vehicles in Dynamic Environments Using Flexible Formation Graphs and Tight Trajectory Hulls”.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"9399-9400"},"PeriodicalIF":3.4,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10844994","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993627","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}
Pub Date : 2025-01-16DOI: 10.1109/ACCESS.2025.3529874
Wei Wu;Wei Luo;Xing Liu;Jia Cui;Pengyu Zhang
This study investigates the motion patterns of the nested test cabin in a gunpowder gas overload test device. Multiple factors during the overload impact process were explored. Under the conditions of keeping the gunpowder combustion model, the friction coefficient between the inner and outer cabins, and the mass of the cabins unchanged, the special acceleration curve and its frequency spectrum and the impact response spectrum of the pseudo-velocity are analyzed. Numerical simulations and experimental studies revealed that there is compound motion between the inner and outer cabins in the gunpowder gas overload test device, resulting in small oscillations (referred to as oscillation wavelets) in the measurement results of the test system within the inner cabin. These oscillation wavelets occur when the critical acceleration of the test cabin reaches approximately 4700g. Increasing the initial velocity of the test cabin leads to a larger amplitude of the oscillation wavelets in the overall acceleration curve. In the frequency domain, secondary spectra appear under high overload conditions, and the bandwidth of the secondary spectra increases with the overload. The PVSRS trends for all overloads are roughly the same, with the variation in peak pseudo velocity being only on the order of $10^{-3}$ . Theoretical analysis and experimental results show good consistency, which helps to better understand the motion process of the test cabin in the gunpowder gas overload test device and provides support for the improvement and design of the overall device.
{"title":"Study on the Motion Patterns of Nested Test Cabin and Its Shock Response Spectrum Analysis","authors":"Wei Wu;Wei Luo;Xing Liu;Jia Cui;Pengyu Zhang","doi":"10.1109/ACCESS.2025.3529874","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3529874","url":null,"abstract":"This study investigates the motion patterns of the nested test cabin in a gunpowder gas overload test device. Multiple factors during the overload impact process were explored. Under the conditions of keeping the gunpowder combustion model, the friction coefficient between the inner and outer cabins, and the mass of the cabins unchanged, the special acceleration curve and its frequency spectrum and the impact response spectrum of the pseudo-velocity are analyzed. Numerical simulations and experimental studies revealed that there is compound motion between the inner and outer cabins in the gunpowder gas overload test device, resulting in small oscillations (referred to as oscillation wavelets) in the measurement results of the test system within the inner cabin. These oscillation wavelets occur when the critical acceleration of the test cabin reaches approximately 4700g. Increasing the initial velocity of the test cabin leads to a larger amplitude of the oscillation wavelets in the overall acceleration curve. In the frequency domain, secondary spectra appear under high overload conditions, and the bandwidth of the secondary spectra increases with the overload. The PVSRS trends for all overloads are roughly the same, with the variation in peak pseudo velocity being only on the order of <inline-formula> <tex-math>$10^{-3}$ </tex-math></inline-formula>. Theoretical analysis and experimental results show good consistency, which helps to better understand the motion process of the test cabin in the gunpowder gas overload test device and provides support for the improvement and design of the overall device.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"12044-12054"},"PeriodicalIF":3.4,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10843187","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993005","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}
Pub Date : 2025-01-16DOI: 10.1109/ACCESS.2025.3529872
Jinjiang Zhang;Qiang Lin;Lu Wang;Orefo Victor Arinze;Zihan Hu;Yantai Huang
Deep reinforcement learning has been widely applied in the field of residential energy management, showcasing considerable promise in enhancing energy efficiency and reducing energy consumption. However, it is observed that some methodologies still suffer from inadequate data exploitation, which results in suboptimal policy performance. In this study, focusing on the residential energy management system, an innovative reinforcement learning method is proposed. This novel method fuses the asynchronous advantage actor-critic architecture with a familiarity-based experience replay mechanism, with the ambition of markedly improving learning efficiency and control performance. Numerical comparisons were made to justify the effectiveness of the method. Experimental results across diverse cases confirm that the proposed algorithm can effectively achieve optimal energy scheduling for residential sectors. Furthermore, the proposed methodology has demonstrated a notable reduction in grid interaction expenses, achieving a decrease of 27.03% and 16.38% relative to the other two scenarios. In comparison with the Proximal Policy Optimization (PPO) and Deep Q-Network (DQN) algorithms, the novel approach not only improves the average reward value post-convergence by 38.48% and 47.17%, respectively, but also significantly reduces the training duration by 81.19% within a multi-threaded computational environment.
{"title":"Residential Energy Management Method Based on the Proposed A3C-FER","authors":"Jinjiang Zhang;Qiang Lin;Lu Wang;Orefo Victor Arinze;Zihan Hu;Yantai Huang","doi":"10.1109/ACCESS.2025.3529872","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3529872","url":null,"abstract":"Deep reinforcement learning has been widely applied in the field of residential energy management, showcasing considerable promise in enhancing energy efficiency and reducing energy consumption. However, it is observed that some methodologies still suffer from inadequate data exploitation, which results in suboptimal policy performance. In this study, focusing on the residential energy management system, an innovative reinforcement learning method is proposed. This novel method fuses the asynchronous advantage actor-critic architecture with a familiarity-based experience replay mechanism, with the ambition of markedly improving learning efficiency and control performance. Numerical comparisons were made to justify the effectiveness of the method. Experimental results across diverse cases confirm that the proposed algorithm can effectively achieve optimal energy scheduling for residential sectors. Furthermore, the proposed methodology has demonstrated a notable reduction in grid interaction expenses, achieving a decrease of 27.03% and 16.38% relative to the other two scenarios. In comparison with the Proximal Policy Optimization (PPO) and Deep Q-Network (DQN) algorithms, the novel approach not only improves the average reward value post-convergence by 38.48% and 47.17%, respectively, but also significantly reduces the training duration by 81.19% within a multi-threaded computational environment.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"12203-12214"},"PeriodicalIF":3.4,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10843226","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993734","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}
Pub Date : 2025-01-15DOI: 10.1109/ACCESS.2025.3530245
Y. V. Nandini;T. Jaya Lakshmi;Murali Krishna Enduri;Mohd Zairul Mazwan Jilani
In complex networks, predicting the formation of new connections, or links, within complex networks has been a central challenge, traditionally addressed using graph-based models. These models, however, are limited in their ability to capture higher-order interactions that exist in many real-world networks, such as social, biological, and technological systems. To account for these multi-node interactions, hyper-networks have emerged as a more flexible framework, where hyperedges can connect multiple nodes simultaneously. Traditional link prediction methods often treat all common neighbors equally, overlooking the fact that not all nodes contribute uniformly to the formation of future links. Each node within a network holds a distinct level of importance, which can influence the likelihood of link formation among its neighbors. To address this, we introduce a link prediction approach leveraging hypercentrality measures adapted from traditional centrality metrics such as degree, clustering coefficient, betweenness, and closeness to capture node significance and improve link prediction in hyper-networks. We propose the Link Prediction Based on HyperCentrality in hyper-networks (LPHC) model, which enhances traditional common neighbor and jaccard coefficient of hyper-network frameworks by incorporating centrality scores to account for node importance. Our approach is evaluated across multiple real-world hyper-networks datasets, demonstrating its superiority over traditional link prediction methods. The results show that link prediction in hypercentrality-based models, particularly those utilizing hyperdegree and hyperclustering coefficients for common neighbor and jaccard coefficent approaches in hyper-networks, consistently outperform existing methods in terms of both F1-score and Area Under the Precision-Recall Curve (AUPR), offering a more precise understanding of potential link formations in hyper-networks. The proposed LPHC model consistently outperforms the existing HCN and HJC models across all datasets, achieving an overall improvement of 69% compared to HCN and 68% compared to HJC.
{"title":"Link Prediction in Complex Hyper-Networks Leveraging HyperCentrality","authors":"Y. V. Nandini;T. Jaya Lakshmi;Murali Krishna Enduri;Mohd Zairul Mazwan Jilani","doi":"10.1109/ACCESS.2025.3530245","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3530245","url":null,"abstract":"In complex networks, predicting the formation of new connections, or links, within complex networks has been a central challenge, traditionally addressed using graph-based models. These models, however, are limited in their ability to capture higher-order interactions that exist in many real-world networks, such as social, biological, and technological systems. To account for these multi-node interactions, hyper-networks have emerged as a more flexible framework, where hyperedges can connect multiple nodes simultaneously. Traditional link prediction methods often treat all common neighbors equally, overlooking the fact that not all nodes contribute uniformly to the formation of future links. Each node within a network holds a distinct level of importance, which can influence the likelihood of link formation among its neighbors. To address this, we introduce a link prediction approach leveraging hypercentrality measures adapted from traditional centrality metrics such as degree, clustering coefficient, betweenness, and closeness to capture node significance and improve link prediction in hyper-networks. We propose the Link Prediction Based on HyperCentrality in hyper-networks (LPHC) model, which enhances traditional common neighbor and jaccard coefficient of hyper-network frameworks by incorporating centrality scores to account for node importance. Our approach is evaluated across multiple real-world hyper-networks datasets, demonstrating its superiority over traditional link prediction methods. The results show that link prediction in hypercentrality-based models, particularly those utilizing hyperdegree and hyperclustering coefficients for common neighbor and jaccard coefficent approaches in hyper-networks, consistently outperform existing methods in terms of both F1-score and Area Under the Precision-Recall Curve (AUPR), offering a more precise understanding of potential link formations in hyper-networks. The proposed LPHC model consistently outperforms the existing HCN and HJC models across all datasets, achieving an overall improvement of 69% compared to HCN and 68% compared to HJC.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"12239-12254"},"PeriodicalIF":3.4,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10843197","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993798","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}
Pub Date : 2025-01-15DOI: 10.1109/ACCESS.2025.3529839
Olanrewaju L. Abraham;Md Asri Bin Ngadi;Johan Bin Mohamad Sharif;Mohd Kufaisal Mohd Sidik
Task scheduling in cloud computing environment aims to identify alternative methods for effectively allocating competing cloud tasks to constrained resources, optimizing one or more objectives. This systematic literature review (SLR) examines advancements in multi-objective optimization techniques for cloud task scheduling from year 2010 to October 2024, providing an up-to-date analysis of the field. Cloud task scheduling, critical for optimizing performance, cost, and resource use, increasingly relies on multi-objective approaches to address complex and competing scheduling goals. This comprehensive review presents a detailed taxonomy and classification of multi-objective optimization methods, highlighting trends and developments across various approaches. Additionally, we conduct a comparative analysis of key scheduling objectives, testing environments, statistical evaluation methods, and datasets employed in recent studies, offering insights into current practices and best-fit approaches for different scenarios. The findings of this SLR aim to guide researchers and practitioners in selecting appropriate techniques, metrics, and datasets, supporting effective decision-making and advancing the design of cloud task scheduling systems.
{"title":"Multi-Objective Optimization Techniques in Cloud Task Scheduling: A Systematic Literature Review","authors":"Olanrewaju L. Abraham;Md Asri Bin Ngadi;Johan Bin Mohamad Sharif;Mohd Kufaisal Mohd Sidik","doi":"10.1109/ACCESS.2025.3529839","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3529839","url":null,"abstract":"Task scheduling in cloud computing environment aims to identify alternative methods for effectively allocating competing cloud tasks to constrained resources, optimizing one or more objectives. This systematic literature review (SLR) examines advancements in multi-objective optimization techniques for cloud task scheduling from year 2010 to October 2024, providing an up-to-date analysis of the field. Cloud task scheduling, critical for optimizing performance, cost, and resource use, increasingly relies on multi-objective approaches to address complex and competing scheduling goals. This comprehensive review presents a detailed taxonomy and classification of multi-objective optimization methods, highlighting trends and developments across various approaches. Additionally, we conduct a comparative analysis of key scheduling objectives, testing environments, statistical evaluation methods, and datasets employed in recent studies, offering insights into current practices and best-fit approaches for different scenarios. The findings of this SLR aim to guide researchers and practitioners in selecting appropriate techniques, metrics, and datasets, supporting effective decision-making and advancing the design of cloud task scheduling systems.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"12255-12291"},"PeriodicalIF":3.4,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10843235","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993201","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}
Pub Date : 2025-01-15DOI: 10.1109/ACCESS.2025.3530091
Yueying Zhou;Junji Jiang;Lijun Wang;Shanshan Liang;Hao Liu
The automatic and effective detection of cognitive load for air traffic control (ATC) operators through electroencephalography (EEG) signals provides a covert and objective method for enhancing ATC safety. Nevertheless, the extant paradigm is limited to simple cognitive tasks and lacks real-world scenarios. In this study, a cognitive load-elicited experiment was therefore designed to record the EEG data of eight ATC operators under four distinct simulation scenarios, ascertaining whether they experienced varying degrees of workload. Subsequently, the collected EEG signal was preprocessed. We then used one hybrid deep learning model based on the convolutional layers and a self-attention mechanism to extract the pertinent EEG features. In conjunction with multi-layer perceptron, we decoded cognitive load state into low, high, overload, and special. The experimental results demonstrated that EEG could serve as a reliable measure for predicting ATC load, with an average accuracy of 88.76% and a peak accuracy of 99% at the single-subject level. Additionally, it highlighted the critical role of the frontal regions in decoding cognitive load. This study serves to enhance the efficacy of personalized EEG decoding for ATC operators, furnishing evidence for the feasibility of developing an intelligent load-detecting system.
{"title":"Enhanced Cognitive Load Detection in Air Traffic Control Operators Using EEG and a Hybrid Deep Learning Approach","authors":"Yueying Zhou;Junji Jiang;Lijun Wang;Shanshan Liang;Hao Liu","doi":"10.1109/ACCESS.2025.3530091","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3530091","url":null,"abstract":"The automatic and effective detection of cognitive load for air traffic control (ATC) operators through electroencephalography (EEG) signals provides a covert and objective method for enhancing ATC safety. Nevertheless, the extant paradigm is limited to simple cognitive tasks and lacks real-world scenarios. In this study, a cognitive load-elicited experiment was therefore designed to record the EEG data of eight ATC operators under four distinct simulation scenarios, ascertaining whether they experienced varying degrees of workload. Subsequently, the collected EEG signal was preprocessed. We then used one hybrid deep learning model based on the convolutional layers and a self-attention mechanism to extract the pertinent EEG features. In conjunction with multi-layer perceptron, we decoded cognitive load state into low, high, overload, and special. The experimental results demonstrated that EEG could serve as a reliable measure for predicting ATC load, with an average accuracy of 88.76% and a peak accuracy of 99% at the single-subject level. Additionally, it highlighted the critical role of the frontal regions in decoding cognitive load. This study serves to enhance the efficacy of personalized EEG decoding for ATC operators, furnishing evidence for the feasibility of developing an intelligent load-detecting system.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"12127-12137"},"PeriodicalIF":3.4,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10843192","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993344","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}
In computer vision, most existing works about object detection focus on detecting objects in the good lighting conditions instead of low-light conditions. Even the few existing works that are centered on object detection in the low-light conditions, predominantly focus on the general object detection rather than the detection of small objects. The main challenges affecting small object detection accuracy in low-light conditions are occlusion caused by the low light, shadows, and darkness that adversely affect the surrounding context leading to poor object classification and the insufficient spatial information that negatively affect object localization resulting in poor small object detection. To address the challenge of poor small object detection in low-light conditions we propose the Hybrid Intersection over Union (HIoU) localization loss to enhance the detection accuracy of small objects in these conditions. This loss utilizes the top-bottom distances of the targeted and predicted bounding boxes and the manhattan distance of the boxes’ centres to deal with the issue of misalignment that negatively affect the small object detection accuracy. Also, it jointly works with the classification loss to offer a joint optimization that facilitates a network to learn features that are important for both localization and classification. Experimental results show that the proposed loss enhances the detection accuracy of small objects in low-light conditions.
{"title":"Hybrid Intersection Over Union Loss for a Robust Small Object Detection in Low-Light Conditions","authors":"Twahir Kiobya;Junfeng Zhou;Baraka Maiseli;Maqbool Khan","doi":"10.1109/ACCESS.2025.3530089","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3530089","url":null,"abstract":"In computer vision, most existing works about object detection focus on detecting objects in the good lighting conditions instead of low-light conditions. Even the few existing works that are centered on object detection in the low-light conditions, predominantly focus on the general object detection rather than the detection of small objects. The main challenges affecting small object detection accuracy in low-light conditions are occlusion caused by the low light, shadows, and darkness that adversely affect the surrounding context leading to poor object classification and the insufficient spatial information that negatively affect object localization resulting in poor small object detection. To address the challenge of poor small object detection in low-light conditions we propose the Hybrid Intersection over Union (HIoU) localization loss to enhance the detection accuracy of small objects in these conditions. This loss utilizes the top-bottom distances of the targeted and predicted bounding boxes and the manhattan distance of the boxes’ centres to deal with the issue of misalignment that negatively affect the small object detection accuracy. Also, it jointly works with the classification loss to offer a joint optimization that facilitates a network to learn features that are important for both localization and classification. Experimental results show that the proposed loss enhances the detection accuracy of small objects in low-light conditions.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"12321-12331"},"PeriodicalIF":3.4,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10843225","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993867","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}
Pub Date : 2025-01-14DOI: 10.1109/ACCESS.2025.3529758
Linlin Liu;Lele Niu;Jun Tang;Yong Ding
Video Super-Resolution (VSR) aims to reconstruct high-quality high-resolution (HR) videos from low-resolution (LR) inputs. Recent studies have explored diffusion models (DMs) for VSR by exploiting their generative priors to produce realistic details. However, the inherent randomness of diffusion models presents significant challenges for controlling content. In particular, current DM-based VSR methods often neglect inter-frame temporal coherence and reconstruction-oriented objectives, leading to visual distortion and temporal inconsistency. In this paper, we introduce VSRDiff, a DM-based framework for VSR that emphasizes inter-frame temporal coherence and adopts a novel reconstruction perspective. Specifically, the Inter-Frame Aggregation Guidance (IFAG) module is developed to learn contextual inter-frame aggregation guidance, alleviating visual distortion caused by the randomness of diffusion models. Furthermore, the Progressive Reconstruction Sampling (PRS) approach is employed to generate reconstruction-oriented latents, balancing fidelity and detail richness. Additionally, temporal consistency is enhanced through second-order bidirectional latent propagation using the Flow-guided Latent Correction (FLC) module. Extensive experiments on the REDS4 and Vid4 datasets demonstrate that VSRDiff achieves highly competitive VSR performance with more realistic details, surpassing existing state-of-the-art methods in both visual fidelity and temporal consistency. Specifically, VSRDiff achieves the best scores on the REDS4 dataset in LPIPS, DISTS, and NIQE, with values of 0.1137, 0.0445, and 2.970, respectively. The result will be released at https://github.com/aigcvsr/VSRDiff.
{"title":"VSRDiff: Learning Inter-Frame Temporal Coherence in Diffusion Model for Video Super-Resolution","authors":"Linlin Liu;Lele Niu;Jun Tang;Yong Ding","doi":"10.1109/ACCESS.2025.3529758","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3529758","url":null,"abstract":"Video Super-Resolution (VSR) aims to reconstruct high-quality high-resolution (HR) videos from low-resolution (LR) inputs. Recent studies have explored diffusion models (DMs) for VSR by exploiting their generative priors to produce realistic details. However, the inherent randomness of diffusion models presents significant challenges for controlling content. In particular, current DM-based VSR methods often neglect inter-frame temporal coherence and reconstruction-oriented objectives, leading to visual distortion and temporal inconsistency. In this paper, we introduce VSRDiff, a DM-based framework for VSR that emphasizes inter-frame temporal coherence and adopts a novel reconstruction perspective. Specifically, the Inter-Frame Aggregation Guidance (IFAG) module is developed to learn contextual inter-frame aggregation guidance, alleviating visual distortion caused by the randomness of diffusion models. Furthermore, the Progressive Reconstruction Sampling (PRS) approach is employed to generate reconstruction-oriented latents, balancing fidelity and detail richness. Additionally, temporal consistency is enhanced through second-order bidirectional latent propagation using the Flow-guided Latent Correction (FLC) module. Extensive experiments on the REDS4 and Vid4 datasets demonstrate that VSRDiff achieves highly competitive VSR performance with more realistic details, surpassing existing state-of-the-art methods in both visual fidelity and temporal consistency. Specifically, VSRDiff achieves the best scores on the REDS4 dataset in LPIPS, DISTS, and NIQE, with values of 0.1137, 0.0445, and 2.970, respectively. The result will be released at <uri>https://github.com/aigcvsr/VSRDiff</uri>.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"11447-11462"},"PeriodicalIF":3.4,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10840194","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993569","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：