Edge deployment solutions based on convolutional neural networks(CNNs)have garnered significant attention because of their potential applications.However,traditional CNNs rely on pooling to reduce the feature size,lea...Edge deployment solutions based on convolutional neural networks(CNNs)have garnered significant attention because of their potential applications.However,traditional CNNs rely on pooling to reduce the feature size,leading to substantial information loss and reduced network robustness.Herein,we propose a more robust adaptive pooling network(APN)method implemented using memristor technology.Our method introduces an improved pooling layer that reduces input features to an arbitrary scale without compromising their importance.Different coupling coefficients of the pooling layer are stored as conductance values in arrays.We validate the proposed APN on generic datasets,demonstrating significant performance improvements over previously reported CNN architectures.Additionally,we evaluate the APN on a CAPTCHA recognition task with perturbations to assess network robustness.The results show that the APN achieves 92.6% accuracy in 4-digit CAPTCHA recognition and exhibits higher robustness.This brief presents a highly robust and novel scheme for edge computing using memristor technology.展开更多
With the rapid development of intelligent video surveillance technology,pedestrian re-identification has become increasingly important inmulti-camera surveillance systems.This technology plays a critical role in enhan...With the rapid development of intelligent video surveillance technology,pedestrian re-identification has become increasingly important inmulti-camera surveillance systems.This technology plays a critical role in enhancing public safety.However,traditional methods typically process images and text separately,applying upstream models directly to downstream tasks.This approach significantly increases the complexity ofmodel training and computational costs.Furthermore,the common class imbalance in existing training datasets limitsmodel performance improvement.To address these challenges,we propose an innovative framework named Person Re-ID Network Based on Visual Prompt Technology andMulti-Instance Negative Pooling(VPM-Net).First,we incorporate the Contrastive Language-Image Pre-training(CLIP)pre-trained model to accurately map visual and textual features into a unified embedding space,effectively mitigating inconsistencies in data distribution and the training process.To enhancemodel adaptability and generalization,we introduce an efficient and task-specific Visual Prompt Tuning(VPT)technique,which improves the model’s relevance to specific tasks.Additionally,we design two key modules:the Knowledge-Aware Network(KAN)and theMulti-Instance Negative Pooling(MINP)module.The KAN module significantly enhances the model’s understanding of complex scenarios through deep contextual semantic modeling.MINP module handles samples,effectively improving the model’s ability to distinguish fine-grained features.The experimental outcomes across diverse datasets underscore the remarkable performance of VPM-Net.These results vividly demonstrate the unique advantages and robust reliability of VPM-Net in fine-grained retrieval tasks.展开更多
This study focuses on tool condition recognition through data-driven approaches to enhance the intelligence level of computerized numerical control(CNC)machining processes and improve tool utilization efficiency.Tradi...This study focuses on tool condition recognition through data-driven approaches to enhance the intelligence level of computerized numerical control(CNC)machining processes and improve tool utilization efficiency.Traditional tool monitoring methods that rely on empirical knowledge or limited mathematical models struggle to adapt to complex and dynamic machining environments.To address this,we implement real-time tool condition recognition by introducing deep learning technology.Aiming to the insufficient recognition accuracy,we propose a pyramid pooling-based vision Transformer network(P2ViT-Net)method for tool condition recognition.Using images as input effectively mitigates the issue of low-dimensional signal features.We enhance the vision Transformer(ViT)framework for image classification by developing the P2ViT model and adapt it to tool condition recognition.Experimental results demonstrate that our improved P2ViT model achieves 94.4%recognition accuracy,showing a 10%improvement over conventional ViT and outperforming all comparative convolutional neural network models.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.62274002,62304001,and 62201005)the Anhui Provincial Natural Science Foundation(Grant Nos.2308085QF213 and 2408085QF211)the Natural Science Research Project of the Anhui Educational Committee(Grant No.2023AH050072)。
文摘Edge deployment solutions based on convolutional neural networks(CNNs)have garnered significant attention because of their potential applications.However,traditional CNNs rely on pooling to reduce the feature size,leading to substantial information loss and reduced network robustness.Herein,we propose a more robust adaptive pooling network(APN)method implemented using memristor technology.Our method introduces an improved pooling layer that reduces input features to an arbitrary scale without compromising their importance.Different coupling coefficients of the pooling layer are stored as conductance values in arrays.We validate the proposed APN on generic datasets,demonstrating significant performance improvements over previously reported CNN architectures.Additionally,we evaluate the APN on a CAPTCHA recognition task with perturbations to assess network robustness.The results show that the APN achieves 92.6% accuracy in 4-digit CAPTCHA recognition and exhibits higher robustness.This brief presents a highly robust and novel scheme for edge computing using memristor technology.
基金funded by the Key Research and Development Program of Hubei Province,China(Grant No.2023BEB024)the Young and Middle-aged Scientific and Technological Innova-tion Team Plan in Higher Education Institutions inHubei Province,China(GrantNo.T2023007)the key projects ofHubei Provincial Department of Education(No.D20161403).
文摘With the rapid development of intelligent video surveillance technology,pedestrian re-identification has become increasingly important inmulti-camera surveillance systems.This technology plays a critical role in enhancing public safety.However,traditional methods typically process images and text separately,applying upstream models directly to downstream tasks.This approach significantly increases the complexity ofmodel training and computational costs.Furthermore,the common class imbalance in existing training datasets limitsmodel performance improvement.To address these challenges,we propose an innovative framework named Person Re-ID Network Based on Visual Prompt Technology andMulti-Instance Negative Pooling(VPM-Net).First,we incorporate the Contrastive Language-Image Pre-training(CLIP)pre-trained model to accurately map visual and textual features into a unified embedding space,effectively mitigating inconsistencies in data distribution and the training process.To enhancemodel adaptability and generalization,we introduce an efficient and task-specific Visual Prompt Tuning(VPT)technique,which improves the model’s relevance to specific tasks.Additionally,we design two key modules:the Knowledge-Aware Network(KAN)and theMulti-Instance Negative Pooling(MINP)module.The KAN module significantly enhances the model’s understanding of complex scenarios through deep contextual semantic modeling.MINP module handles samples,effectively improving the model’s ability to distinguish fine-grained features.The experimental outcomes across diverse datasets underscore the remarkable performance of VPM-Net.These results vividly demonstrate the unique advantages and robust reliability of VPM-Net in fine-grained retrieval tasks.
基金supported by China Postdoctoral Science Foundation(No.2024M754122)the Postdoctoral Fellowship Programof CPSF(No.GZB20240972)+3 种基金the Jiangsu Funding Program for Excellent Postdoctoral Talent(No.2024ZB194)Natural Science Foundation of Jiangsu Province(No.BK20241389)Basic Science ResearchFund of China(No.JCKY2023203C026)2024 Jiangsu Province Talent Programme Qinglan Project.
文摘This study focuses on tool condition recognition through data-driven approaches to enhance the intelligence level of computerized numerical control(CNC)machining processes and improve tool utilization efficiency.Traditional tool monitoring methods that rely on empirical knowledge or limited mathematical models struggle to adapt to complex and dynamic machining environments.To address this,we implement real-time tool condition recognition by introducing deep learning technology.Aiming to the insufficient recognition accuracy,we propose a pyramid pooling-based vision Transformer network(P2ViT-Net)method for tool condition recognition.Using images as input effectively mitigates the issue of low-dimensional signal features.We enhance the vision Transformer(ViT)framework for image classification by developing the P2ViT model and adapt it to tool condition recognition.Experimental results demonstrate that our improved P2ViT model achieves 94.4%recognition accuracy,showing a 10%improvement over conventional ViT and outperforming all comparative convolutional neural network models.
