Pub Date : 2024-07-18DOI: 10.3390/electronics13142820
Alpaslan Gökcen, Ali Boyacı
This study introduces a privacy-preserving approach for the real-time action detection in intelligent vehicles using a federated learning (FL)-based temporal recurrent network (TRN). This approach enables edge devices to independently train models, enhancing data privacy and scalability by eliminating central data consolidation. Our FL-based TRN effectively captures temporal dependencies, anticipating future actions with high precision. Extensive testing on the Honda HDD and TVSeries datasets demonstrated robust performance in centralized and decentralized settings, with competitive mean average precision (mAP) scores. The experimental results highlighted that our FL-based TRN achieved an mAP of 40.0% in decentralized settings, closely matching the 40.1% in centralized configurations. Notably, the model excelled in detecting complex driving maneuvers, with mAPs of 80.7% for intersection passing and 78.1% for right turns. These outcomes affirm the model’s accuracy in action localization and identification. The system showed significant scalability and adaptability, maintaining robust performance across increased client device counts. The integration of a temporal decoder enabled predictions of future actions up to 2 s ahead, enhancing the responsiveness. Our research advances intelligent vehicle technology, promoting safety and efficiency while maintaining strict privacy standards.
{"title":"Privacy-Preserving Real-Time Action Detection in Intelligent Vehicles Using Federated Learning-Based Temporal Recurrent Network","authors":"Alpaslan Gökcen, Ali Boyacı","doi":"10.3390/electronics13142820","DOIUrl":"https://doi.org/10.3390/electronics13142820","url":null,"abstract":"This study introduces a privacy-preserving approach for the real-time action detection in intelligent vehicles using a federated learning (FL)-based temporal recurrent network (TRN). This approach enables edge devices to independently train models, enhancing data privacy and scalability by eliminating central data consolidation. Our FL-based TRN effectively captures temporal dependencies, anticipating future actions with high precision. Extensive testing on the Honda HDD and TVSeries datasets demonstrated robust performance in centralized and decentralized settings, with competitive mean average precision (mAP) scores. The experimental results highlighted that our FL-based TRN achieved an mAP of 40.0% in decentralized settings, closely matching the 40.1% in centralized configurations. Notably, the model excelled in detecting complex driving maneuvers, with mAPs of 80.7% for intersection passing and 78.1% for right turns. These outcomes affirm the model’s accuracy in action localization and identification. The system showed significant scalability and adaptability, maintaining robust performance across increased client device counts. The integration of a temporal decoder enabled predictions of future actions up to 2 s ahead, enhancing the responsiveness. Our research advances intelligent vehicle technology, promoting safety and efficiency while maintaining strict privacy standards.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 21","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141826904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142825
Feng Zhou, Zhuxuan Cheng, Haitao Yang, Yifeng Song, Shengpeng Fu
The visible-infrared person re-identification (VI-ReID) task aims to retrieve the same pedestrian between visible and infrared images. VI-ReID is a challenging task due to the huge modality discrepancy and complex intra-modality variations. Existing works mainly complete the modality alignment at one stage. However, aligning modalities at different stages has positive effects on the intra-class and inter-class distances of cross-modality features, which are often ignored. Moreover, discriminative features with identity information may be corrupted in the processing of modality alignment, further degrading the performance of person re-identification. In this paper, we propose a progressive discriminative feature learning (PDFL) network that adopts different alignment strategies at different stages to alleviate the discrepancy and learn discriminative features progressively. Specifically, we first design an adaptive cross fusion module (ACFM) to learn the identity-relevant features via modality alignment with channel-level attention. For well preserving identity information, we propose a dual-attention-guided instance normalization module (DINM), which can well guide instance normalization to align two modalities into a unified feature space through channel and spatial information embedding. Finally, we generate multiple part features of a person to mine subtle differences. Multi-loss optimization is imposed during the training process for more effective learning supervision. Extensive experiments on the public datasets of SYSU-MM01 and RegDB validate that our proposed method performs favorably against most state-of-the-art methods.
{"title":"Progressive Discriminative Feature Learning for Visible-Infrared Person Re-Identification","authors":"Feng Zhou, Zhuxuan Cheng, Haitao Yang, Yifeng Song, Shengpeng Fu","doi":"10.3390/electronics13142825","DOIUrl":"https://doi.org/10.3390/electronics13142825","url":null,"abstract":"The visible-infrared person re-identification (VI-ReID) task aims to retrieve the same pedestrian between visible and infrared images. VI-ReID is a challenging task due to the huge modality discrepancy and complex intra-modality variations. Existing works mainly complete the modality alignment at one stage. However, aligning modalities at different stages has positive effects on the intra-class and inter-class distances of cross-modality features, which are often ignored. Moreover, discriminative features with identity information may be corrupted in the processing of modality alignment, further degrading the performance of person re-identification. In this paper, we propose a progressive discriminative feature learning (PDFL) network that adopts different alignment strategies at different stages to alleviate the discrepancy and learn discriminative features progressively. Specifically, we first design an adaptive cross fusion module (ACFM) to learn the identity-relevant features via modality alignment with channel-level attention. For well preserving identity information, we propose a dual-attention-guided instance normalization module (DINM), which can well guide instance normalization to align two modalities into a unified feature space through channel and spatial information embedding. Finally, we generate multiple part features of a person to mine subtle differences. Multi-loss optimization is imposed during the training process for more effective learning supervision. Extensive experiments on the public datasets of SYSU-MM01 and RegDB validate that our proposed method performs favorably against most state-of-the-art methods.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 47","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141824142","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142830
Jiabao Wang, Chao Guo, Wanyu Zhou, Qin Wan
The electrically excited homopolar inductor machine has a static excitation coil as well as a robust rotor, which makes it attractive in the field of high-speed superconducting machines. This paper designed and analyzed a megawatt class superconducting homopolar inductor machine for aerospace application. To improve the power density, a mass-reduced rotor structure is proposed. Firstly, the main structure parameters of the superconducting homopolar inductor machine are derived based on the required power and speed. Secondly, the electromagnetic performance of the superconducting homopolar inductor machine is analyzed based on the finite element method. Thirdly, a mass-reduced rotor is proposed to improve its power density. The structural performance of the rotor and the electromagnetic performance of the superconducting homopolar inductor machine before and after rotor-mass reduction are evaluated. Compared with the initial rotor, the mass of the mass-reduced rotor is reduced from 66.56 kg to 50.02 kg, which increases the power density by 14.3%. The result shows that a superconducting homopolar inductor machine with a mass-reduced rotor can effectively improve its power density without affecting its output power.
{"title":"Design and Analysis of a Superconducting Homopolar Inductor Machine for Aerospace Application","authors":"Jiabao Wang, Chao Guo, Wanyu Zhou, Qin Wan","doi":"10.3390/electronics13142830","DOIUrl":"https://doi.org/10.3390/electronics13142830","url":null,"abstract":"The electrically excited homopolar inductor machine has a static excitation coil as well as a robust rotor, which makes it attractive in the field of high-speed superconducting machines. This paper designed and analyzed a megawatt class superconducting homopolar inductor machine for aerospace application. To improve the power density, a mass-reduced rotor structure is proposed. Firstly, the main structure parameters of the superconducting homopolar inductor machine are derived based on the required power and speed. Secondly, the electromagnetic performance of the superconducting homopolar inductor machine is analyzed based on the finite element method. Thirdly, a mass-reduced rotor is proposed to improve its power density. The structural performance of the rotor and the electromagnetic performance of the superconducting homopolar inductor machine before and after rotor-mass reduction are evaluated. Compared with the initial rotor, the mass of the mass-reduced rotor is reduced from 66.56 kg to 50.02 kg, which increases the power density by 14.3%. The result shows that a superconducting homopolar inductor machine with a mass-reduced rotor can effectively improve its power density without affecting its output power.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 83","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141827024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142824
Ming Huang, Xia Shang, Xiang Chen, Feng Zhang, Bing Li, Baojun Lan, Shuang Chen, Jun Zhu
The main functions of the space information system, such as providing the backbone transmission, broadband access, and global connectivity, are realized based on the network topology. Thus, it is necessary to recognize the temporal dynamics of the network topology. A temporal continuity expression method is proposed to describe the topological dynamic characteristics of the network in space information systems. Based on orbit dynamics, a time-dependent adjacency matrix of the space information system can be established by introducing the geometric linkable factor, the link distance intensity factor, and the relative angular velocity factor of the node. The adjacency matrix describes the dynamic characteristics from two layers: one is the physical layer using a time-dependent function, which represents the feasibility of inter-satellite link construction in the system cycle; the other one is the transport layer, described by a piecewise continuous function that varies with time, which characterizes the link quality during the connection period between two satellites. The results show that compared with the existing network topology description methods, the proposed method describes the network topology more accurately, which can distinguish the network topology characteristics at any time, and is more conducive to the understanding and application of the network topology of the space information system.
{"title":"Temporal Continuity Expression for Network Topology of Space Information Systems","authors":"Ming Huang, Xia Shang, Xiang Chen, Feng Zhang, Bing Li, Baojun Lan, Shuang Chen, Jun Zhu","doi":"10.3390/electronics13142824","DOIUrl":"https://doi.org/10.3390/electronics13142824","url":null,"abstract":"The main functions of the space information system, such as providing the backbone transmission, broadband access, and global connectivity, are realized based on the network topology. Thus, it is necessary to recognize the temporal dynamics of the network topology. A temporal continuity expression method is proposed to describe the topological dynamic characteristics of the network in space information systems. Based on orbit dynamics, a time-dependent adjacency matrix of the space information system can be established by introducing the geometric linkable factor, the link distance intensity factor, and the relative angular velocity factor of the node. The adjacency matrix describes the dynamic characteristics from two layers: one is the physical layer using a time-dependent function, which represents the feasibility of inter-satellite link construction in the system cycle; the other one is the transport layer, described by a piecewise continuous function that varies with time, which characterizes the link quality during the connection period between two satellites. The results show that compared with the existing network topology description methods, the proposed method describes the network topology more accurately, which can distinguish the network topology characteristics at any time, and is more conducive to the understanding and application of the network topology of the space information system.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 39","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141827490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Timely and accurate rainfall prediction is crucial to social life and economic activities. Because of the influence of numerous factors on rainfall, making precise predictions is challenging. In this study, the northern Xinjiang region of China is selected as the research area. Based on the pattern of rainfall in the local area and the needs of real life, rainfall is divided into four levels, namely ‘no rain’, ‘light rain’, ‘moderate rain’, and ‘heavy rain and above’, for rainfall levels nowcasting. To solve the problem that the existing model can only extract a single time dependence and cause the loss of some valuable information in rainfall data, a prediction model named DFFNet, which is based on dual-branch feature fusion, is proposed in this paper. The two branches of the model are composed of Transformer and CNN, which are used to extract time dependence and feature interaction in meteorological data, respectively. The features extracted from the two branches are fused for prediction. To verify the performance of DFFNet, the India public rainfall dataset and some sub-datasets in the UEA dataset are chosen for comparison. Compared with the baseline models, DFFNet achieves the best prediction performance on all the selected datasets; compared with the single-branch model, the training time consumption of DFFNet on the two rainfall datasets is reduced by 21% and 9.6%, respectively, and it has a faster convergence speed. The experimental results show that it has certain theoretical value and application value for the study of rainfall nowcasting.
{"title":"DFFNet: A Rainfall Nowcasting Model Based on Dual-Branch Feature Fusion","authors":"Shuxian Liu, Yulong Liu, Jiong Zheng, Yuanyuan Liao, Guohong Zheng, Yongjun Zhang","doi":"10.3390/electronics13142826","DOIUrl":"https://doi.org/10.3390/electronics13142826","url":null,"abstract":"Timely and accurate rainfall prediction is crucial to social life and economic activities. Because of the influence of numerous factors on rainfall, making precise predictions is challenging. In this study, the northern Xinjiang region of China is selected as the research area. Based on the pattern of rainfall in the local area and the needs of real life, rainfall is divided into four levels, namely ‘no rain’, ‘light rain’, ‘moderate rain’, and ‘heavy rain and above’, for rainfall levels nowcasting. To solve the problem that the existing model can only extract a single time dependence and cause the loss of some valuable information in rainfall data, a prediction model named DFFNet, which is based on dual-branch feature fusion, is proposed in this paper. The two branches of the model are composed of Transformer and CNN, which are used to extract time dependence and feature interaction in meteorological data, respectively. The features extracted from the two branches are fused for prediction. To verify the performance of DFFNet, the India public rainfall dataset and some sub-datasets in the UEA dataset are chosen for comparison. Compared with the baseline models, DFFNet achieves the best prediction performance on all the selected datasets; compared with the single-branch model, the training time consumption of DFFNet on the two rainfall datasets is reduced by 21% and 9.6%, respectively, and it has a faster convergence speed. The experimental results show that it has certain theoretical value and application value for the study of rainfall nowcasting.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 45","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141824900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142832
Nut Aroon, Vicky Liu, Luke Kane, Yuefeng Li, A. D. Tesfamicael, Matthew McKague
Attacks launched from IoT networks can cause significant damage to critical network systems and services. IoT networks may contain a large volume of devices. Protecting these devices from being abused to launch traffic amplification attacks is critical. The manufacturer usage description (MUD) architecture uses pre-defined stateless access control rules to allow or block specific network traffic without stateful communication inspection. This can lead to false negative filtering of malicious traffic, as the MUD architecture does not include the monitoring of communication states to determine which connections to allow through. This study presents a novel solution, the enhanced profiling assurance (EPA) architecture. It incorporates both stateless and stateful communication inspection, a unique approach that enhances the detection effectiveness of the MUD architecture. EPA contains layered intrusion detection and prevention systems to monitor stateful and stateless communication. It adopts three-way decision theory with three outcomes: allow, deny, and uncertain. Packets that are marked as uncertain must be continuously monitored to determine access permission. Our analysis, conducted with two network scenarios, demonstrates the superiority of the EPA over the MUD architecture in detecting malicious activities.
{"title":"An Architecture of Enhanced Profiling Assurance for IoT Networks","authors":"Nut Aroon, Vicky Liu, Luke Kane, Yuefeng Li, A. D. Tesfamicael, Matthew McKague","doi":"10.3390/electronics13142832","DOIUrl":"https://doi.org/10.3390/electronics13142832","url":null,"abstract":"Attacks launched from IoT networks can cause significant damage to critical network systems and services. IoT networks may contain a large volume of devices. Protecting these devices from being abused to launch traffic amplification attacks is critical. The manufacturer usage description (MUD) architecture uses pre-defined stateless access control rules to allow or block specific network traffic without stateful communication inspection. This can lead to false negative filtering of malicious traffic, as the MUD architecture does not include the monitoring of communication states to determine which connections to allow through. This study presents a novel solution, the enhanced profiling assurance (EPA) architecture. It incorporates both stateless and stateful communication inspection, a unique approach that enhances the detection effectiveness of the MUD architecture. EPA contains layered intrusion detection and prevention systems to monitor stateful and stateless communication. It adopts three-way decision theory with three outcomes: allow, deny, and uncertain. Packets that are marked as uncertain must be continuously monitored to determine access permission. Our analysis, conducted with two network scenarios, demonstrates the superiority of the EPA over the MUD architecture in detecting malicious activities.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141826460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142837
Ji-Hoon Kwon, Hyeong-Jun Kim, Suk Lee
This study investigates the optimization of traffic scheduling in autonomous vehicle networks using time-sensitive networking (TSN), a type of deterministic Ethernet. Ethernet has high bandwidth and compatibility to support various protocols, and its application range is expanding from office environments to smart factories, aerospace, and automobiles. TSN is a representative technology of deterministic Ethernet and is composed of various standards such as time synchronization, stream reservation, seamless redundancy, frame preemption, and scheduled traffic, which are sub-standards of IEEE 802.1 Ethernet established by the IEEE TSN task group. In order to ensure real-time transmission by minimizing end-to-end delay in a TSN network environment, it is necessary to schedule transmission timing in all links transmitting ST (Scheduled Traffic). This paper proposes network performance metrics and methods for applying machine learning (ML) techniques to optimize traffic scheduling. This study demonstrates that the traffic scheduling problem, which has NP-hard complexity, can be optimized using ML algorithms. The performance of each algorithm is compared and analyzed to identify the scheduling algorithm that best meets the network requirements. Reinforcement learning algorithms, specifically DQN (Deep Q Network) and A2C (Advantage Actor-Critic) were used, and normalized performance metrics (E2E delay, jitter, and guard band bandwidth usage) along with an evaluation function based on their weighted sum were proposed. The performance of each algorithm was evaluated using the topology of a real autonomous vehicle network, and their strengths and weaknesses were compared. The results confirm that artificial intelligence-based algorithms are effective for optimizing TSN traffic scheduling. This study suggests that further theoretical and practical research is needed to enhance the feasibility of applying deterministic Ethernet to autonomous vehicle networks, focusing on time synchronization and schedule optimization.
{"title":"Optimizing Traffic Scheduling in Autonomous Vehicle Networks Using Machine Learning Techniques and Time-Sensitive Networking","authors":"Ji-Hoon Kwon, Hyeong-Jun Kim, Suk Lee","doi":"10.3390/electronics13142837","DOIUrl":"https://doi.org/10.3390/electronics13142837","url":null,"abstract":"This study investigates the optimization of traffic scheduling in autonomous vehicle networks using time-sensitive networking (TSN), a type of deterministic Ethernet. Ethernet has high bandwidth and compatibility to support various protocols, and its application range is expanding from office environments to smart factories, aerospace, and automobiles. TSN is a representative technology of deterministic Ethernet and is composed of various standards such as time synchronization, stream reservation, seamless redundancy, frame preemption, and scheduled traffic, which are sub-standards of IEEE 802.1 Ethernet established by the IEEE TSN task group. In order to ensure real-time transmission by minimizing end-to-end delay in a TSN network environment, it is necessary to schedule transmission timing in all links transmitting ST (Scheduled Traffic). This paper proposes network performance metrics and methods for applying machine learning (ML) techniques to optimize traffic scheduling. This study demonstrates that the traffic scheduling problem, which has NP-hard complexity, can be optimized using ML algorithms. The performance of each algorithm is compared and analyzed to identify the scheduling algorithm that best meets the network requirements. Reinforcement learning algorithms, specifically DQN (Deep Q Network) and A2C (Advantage Actor-Critic) were used, and normalized performance metrics (E2E delay, jitter, and guard band bandwidth usage) along with an evaluation function based on their weighted sum were proposed. The performance of each algorithm was evaluated using the topology of a real autonomous vehicle network, and their strengths and weaknesses were compared. The results confirm that artificial intelligence-based algorithms are effective for optimizing TSN traffic scheduling. This study suggests that further theoretical and practical research is needed to enhance the feasibility of applying deterministic Ethernet to autonomous vehicle networks, focusing on time synchronization and schedule optimization.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 16","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141827805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142839
Xuhua Zhao, Chao Yang, Donglin Zhu, Yujia Liu
To improve the performance of the sparrow search algorithm in solving complex optimization problems, this study proposes a novel variant called the Improved Beetle Antennae Search-Based Sparrow Search Algorithm (IBSSA). A new elite dynamic opposite learning strategy is proposed in the population initialization stage to enhance population diversity. In the update stage of the discoverer, a staged inertia weight guidance mechanism is used to improve the update formula of the discoverer, promote the information exchange between individuals, and improve the algorithm’s ability to optimize on a global level. After the follower’s position is updated, the logarithmic spiral opposition-based learning strategy is introduced to disturb the initial position of the individual in the beetle antennae search algorithm to obtain a more purposeful solution. To address the issue of decreased diversity and susceptibility to local optima in the sparrow population during later stages, the improved beetle antennae search algorithm and sparrow search algorithm are combined using a greedy strategy. This integration aims to improve convergence accuracy. On 20 benchmark test functions and the CEC2017 Test suite, IBSSA performed better than other advanced algorithms. Moreover, six engineering optimization problems were used to demonstrate the improved algorithm’s effectiveness and feasibility.
{"title":"A Hybrid Algorithm Based on Multi-Strategy Elite Learning for Global Optimization","authors":"Xuhua Zhao, Chao Yang, Donglin Zhu, Yujia Liu","doi":"10.3390/electronics13142839","DOIUrl":"https://doi.org/10.3390/electronics13142839","url":null,"abstract":"To improve the performance of the sparrow search algorithm in solving complex optimization problems, this study proposes a novel variant called the Improved Beetle Antennae Search-Based Sparrow Search Algorithm (IBSSA). A new elite dynamic opposite learning strategy is proposed in the population initialization stage to enhance population diversity. In the update stage of the discoverer, a staged inertia weight guidance mechanism is used to improve the update formula of the discoverer, promote the information exchange between individuals, and improve the algorithm’s ability to optimize on a global level. After the follower’s position is updated, the logarithmic spiral opposition-based learning strategy is introduced to disturb the initial position of the individual in the beetle antennae search algorithm to obtain a more purposeful solution. To address the issue of decreased diversity and susceptibility to local optima in the sparrow population during later stages, the improved beetle antennae search algorithm and sparrow search algorithm are combined using a greedy strategy. This integration aims to improve convergence accuracy. On 20 benchmark test functions and the CEC2017 Test suite, IBSSA performed better than other advanced algorithms. Moreover, six engineering optimization problems were used to demonstrate the improved algorithm’s effectiveness and feasibility.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141824380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142818
Athanasios Kanavos, Orestis Papadimitriou, Khalil Al-Hussaeni, Manolis Maragoudakis, Ioannis Karamitsos
White blood cell (WBC) classification is pivotal in medical image analysis, playing a critical role in the precise diagnosis and monitoring of diseases. This paper presents a novel convolutional neural network (CNN) architecture designed specifically for the classification of WBC images. Our model, trained on an extensive dataset, automates the extraction of discriminative features essential for accurate subtype identification. We conducted comprehensive experiments on a publicly available image dataset to validate the efficacy of our methodology. Comparative analysis with state-of-the-art methods shows that our approach significantly outperforms existing models in accurately categorizing WBCs into their respective subtypes. An in-depth analysis of the features learned by the CNN reveals key insights into the morphological traits—such as shape, size, and texture—that contribute to its classification accuracy. Importantly, the model demonstrates robust generalization capabilities, suggesting its high potential for real-world clinical implementation. Our findings indicate that the proposed CNN architecture can substantially enhance the precision and efficiency of WBC subtype identification, offering significant improvements in medical diagnostics and patient care.
{"title":"Advanced Convolutional Neural Networks for Precise White Blood Cell Subtype Classification in Medical Diagnostics","authors":"Athanasios Kanavos, Orestis Papadimitriou, Khalil Al-Hussaeni, Manolis Maragoudakis, Ioannis Karamitsos","doi":"10.3390/electronics13142818","DOIUrl":"https://doi.org/10.3390/electronics13142818","url":null,"abstract":"White blood cell (WBC) classification is pivotal in medical image analysis, playing a critical role in the precise diagnosis and monitoring of diseases. This paper presents a novel convolutional neural network (CNN) architecture designed specifically for the classification of WBC images. Our model, trained on an extensive dataset, automates the extraction of discriminative features essential for accurate subtype identification. We conducted comprehensive experiments on a publicly available image dataset to validate the efficacy of our methodology. Comparative analysis with state-of-the-art methods shows that our approach significantly outperforms existing models in accurately categorizing WBCs into their respective subtypes. An in-depth analysis of the features learned by the CNN reveals key insights into the morphological traits—such as shape, size, and texture—that contribute to its classification accuracy. Importantly, the model demonstrates robust generalization capabilities, suggesting its high potential for real-world clinical implementation. Our findings indicate that the proposed CNN architecture can substantially enhance the precision and efficiency of WBC subtype identification, offering significant improvements in medical diagnostics and patient care.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141824629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-18DOI: 10.3390/electronics13142836
Yanpu Yin, Jiahui Lei, Wei Tao
High-throughput liquid handling workstations are required to process large numbers of test samples in the fields of life sciences and medicine. Liquid retention and droplets hanging in the pipette tips can lead to cross-contamination of samples and reagents and inaccurate experimental results. Traditional methods for detecting liquid retention have low precision and poor real-time performance. This paper proposes an improved YOLOv8 (You Only Look Once version 8) object detection algorithm to address the challenges posed by different liquid sizes and colors, complex situation of test tube racks and multiple samples in the background, and poor global image structure understanding in pipette tip liquid retention detection. A global context (GC) attention mechanism module is introduced into the backbone network and the cross-stage partial feature fusion (C2f) module to better focus on target features. To enhance the ability to effectively combine and process different types of data inputs and background information, a Large Kernel Selection (LKS) module is also introduced into the backbone network. Additionally, the neck network is redesigned to incorporate the Simple Attention (SimAM) mechanism module, generating attention weights and improving overall performance. We evaluated the algorithm using a self-built dataset of pipette tips. Compared to the original YOLOv8 model, the improved algorithm increased mAP@0.5 (mean average precision), F1 score, and precision by 1.7%, 2%, and 1.7%, respectively. The improved YOLOv8 algorithm can enhance the detection capability of liquid-retaining pipette tips, and prevent cross-contamination from affecting the results of sample solution experiments. It provides a detection basis for subsequent automatic processing of solution for liquid retention.
生命科学和医学领域需要高通量液体处理工作站来处理大量测试样品。移液器吸头中的液体滞留和液滴悬挂会导致样品和试剂的交叉污染以及不准确的实验结果。传统的液体滞留检测方法精度低、实时性差。本文提出了一种改进的 YOLOv8（You Only Look Once version 8）对象检测算法，以解决移液管吸头液体滞留检测中不同液体大小和颜色、试管架和背景中多个样品的复杂情况以及全局图像结构理解能力差所带来的挑战。在骨干网络和跨阶段部分特征融合（C2f）模块中引入了全局上下文（GC）关注机制模块，以更好地关注目标特征。为了提高有效组合和处理不同类型数据输入和背景信息的能力，主干网络还引入了大核选择（LKS）模块。此外，我们还重新设计了颈部网络，将简单注意力（SimAM）机制模块纳入其中，以生成注意力权重并提高整体性能。我们使用自建的移液器吸头数据集对算法进行了评估。与最初的 YOLOv8 模型相比，改进后的算法在 mAP@0.5（平均精度）、F1 分数和精度方面分别提高了 1.7%、2% 和 1.7%。改进后的 YOLOv8 算法可以提高留液吸头的检测能力，防止交叉污染影响样品溶液实验结果。它为后续自动处理留液溶液提供了检测依据。
{"title":"Detection of Liquid Retention on Pipette Tips in High-Throughput Liquid Handling Workstations Based on Improved YOLOv8 Algorithm with Attention Mechanism","authors":"Yanpu Yin, Jiahui Lei, Wei Tao","doi":"10.3390/electronics13142836","DOIUrl":"https://doi.org/10.3390/electronics13142836","url":null,"abstract":"High-throughput liquid handling workstations are required to process large numbers of test samples in the fields of life sciences and medicine. Liquid retention and droplets hanging in the pipette tips can lead to cross-contamination of samples and reagents and inaccurate experimental results. Traditional methods for detecting liquid retention have low precision and poor real-time performance. This paper proposes an improved YOLOv8 (You Only Look Once version 8) object detection algorithm to address the challenges posed by different liquid sizes and colors, complex situation of test tube racks and multiple samples in the background, and poor global image structure understanding in pipette tip liquid retention detection. A global context (GC) attention mechanism module is introduced into the backbone network and the cross-stage partial feature fusion (C2f) module to better focus on target features. To enhance the ability to effectively combine and process different types of data inputs and background information, a Large Kernel Selection (LKS) module is also introduced into the backbone network. Additionally, the neck network is redesigned to incorporate the Simple Attention (SimAM) mechanism module, generating attention weights and improving overall performance. We evaluated the algorithm using a self-built dataset of pipette tips. Compared to the original YOLOv8 model, the improved algorithm increased mAP@0.5 (mean average precision), F1 score, and precision by 1.7%, 2%, and 1.7%, respectively. The improved YOLOv8 algorithm can enhance the detection capability of liquid-retaining pipette tips, and prevent cross-contamination from affecting the results of sample solution experiments. It provides a detection basis for subsequent automatic processing of solution for liquid retention.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":" 92","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141824836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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：