Pedestrian detection has been a hot spot in computer vision over the past decades due to the wide spectrum of promising applications,and the major challenge is false positives that occur during pedestrian detection.Th...Pedestrian detection has been a hot spot in computer vision over the past decades due to the wide spectrum of promising applications,and the major challenge is false positives that occur during pedestrian detection.The emergence of various Convolutional Neural Network-based detection strategies substantially enhances pedestrian detection accuracy but still does not solve this problem well.This paper deeply analyzes the detection framework of the two-stage CNN detection methods and finds out false positives in detection results are due to its training strategy misclassifying some false proposals,thus weakening the classification capability of the following subnetwork and hardly suppressing false ones.To solve this problem,this paper proposes a pedestrian-sensitive training algorithm to help two-stage CNN detection methods effectively learn to distinguish the pedestrian and non-pedestrian samples and suppress the false positives in the final detection results.The core of the proposed algorithm is to redesign the training proposal generating scheme for the two-stage CNN detection methods,which can avoid a certain number of false ones that mislead its training process.With the help of the proposed algorithm,the detection accuracy of the MetroNext,a smaller and more accurate metro passenger detector,is further improved,which further decreases false ones in its metro passenger detection results.Based on various challenging benchmark datasets,experiment results have demonstrated that the feasibility of the proposed algorithm is effective in improving pedestrian detection accuracy by removing false positives.Compared with the existing state-of-the-art detection networks,PSTNet demonstrates better overall prediction performance in accuracy,total number of parameters,and inference time;thus,it can become a practical solution for hunting pedestrians on various hardware platforms,especially for mobile and edge devices.展开更多
In recent years,deeps learning has been widely applied in synthetic aperture radar(SAR)image processing.However,the collection of large-scale labeled SAR images is challenging and costly,and the classification accurac...In recent years,deeps learning has been widely applied in synthetic aperture radar(SAR)image processing.However,the collection of large-scale labeled SAR images is challenging and costly,and the classification accuracy is often poor when only limited SAR images are available.To address this issue,we propose a novel framework for sparse SAR target classification under few-shot cases,termed the transfer learning-based interpretable lightweight convolutional neural network(TL-IL-CNN).Additionally,we employ enhanced gradient-weighted class activation mapping(Grad-CAM)to mitigate the“black box”effect often associated with deep learning models and to explore the mechanisms by which a CNN classifies various sparse SAR targets.Initially,we apply a novel bidirectional iterative soft thresholding(BiIST)algorithm to generate sparse images of superior quality compared to those produced by traditional matched filtering(MF)techniques.Subsequently,we pretrain multiple shallow CNNs on a simulated SAR image dataset.Using the sparse SAR dataset as input for the CNNs,we assess the efficacy of transfer learning in sparse SAR target classification and suggest the integration of TL-IL-CNN to enhance the classification accuracy further.Finally,Grad-CAM is utilized to provide visual explanations for the predictions made by the classification framework.The experimental results on the MSTAR dataset reveal that the proposed TL-IL-CNN achieves nearly 90%classification accuracy with only 20%of the training data required under standard operating conditions(SOC),surpassing typical deep learning methods such as vision Transformer(ViT)in the context of small samples.Remarkably,it even presents better performance under extended operating conditions(EOC).Furthermore,the application of Grad-CAM elucidates the CNN’s differentiation process among various sparse SAR targets.The experiments indicate that the model focuses on the target and the background can differ among target classes.The study contributes to an enhanced understanding of the interpretability of such results and enables us to infer the classification outcomes for each category more accurately.展开更多
In this paper,we consider the maximal positive definite solution of the nonlinear matrix equation.By using the idea of Algorithm 2.1 in ZHANG(2013),a new inversion-free method with a stepsize parameter is proposed to ...In this paper,we consider the maximal positive definite solution of the nonlinear matrix equation.By using the idea of Algorithm 2.1 in ZHANG(2013),a new inversion-free method with a stepsize parameter is proposed to obtain the maximal positive definite solution of nonlinear matrix equation X+A^(*)X|^(-α)A=Q with the case 0<α≤1.Based on this method,a new iterative algorithm is developed,and its convergence proof is given.Finally,two numerical examples are provided to show the effectiveness of the proposed method.展开更多
Convex feasibility problems are widely used in image reconstruction,sparse signal recovery,and other areas.This paper is devoted to considering a class of convex feasibility problem arising from sparse signal recovery...Convex feasibility problems are widely used in image reconstruction,sparse signal recovery,and other areas.This paper is devoted to considering a class of convex feasibility problem arising from sparse signal recovery.We rst derive the projection formulas for a vector onto the feasible sets.The centralized circumcentered-reection method is designed to solve the convex feasibility problem.Some numerical experiments demonstrate the feasibility and e ectiveness of the proposed algorithm,showing superior performance compared to conventional alternating projection methods.展开更多
Marine thin plates are susceptible to welding deformation owing to their low structural stiffness.Therefore,the efficient and accurate prediction of welding deformation is essential for improving welding quality.The t...Marine thin plates are susceptible to welding deformation owing to their low structural stiffness.Therefore,the efficient and accurate prediction of welding deformation is essential for improving welding quality.The traditional thermal elastic-plastic finite element method(TEP-FEM)can accurately predict welding deformation.However,its efficiency is low because of the complex nonlinear transient computation,making it difficult to meet the needs of rapid engineering evaluation.To address this challenge,this study proposes an efficient prediction method for welding deformation in marine thin plate butt welds.This method is based on the coupled temperature gradient-thermal strain method(TG-TSM)that integrates inherent strain theory with a shell element finite element model.The proposed method first extracts the distribution pattern and characteristic value of welding-induced inherent strain through TEP-FEM analysis.This strain is then converted into the equivalent thermal load applied to the shell element model for rapid computation.The proposed method-particularly,the gradual temperature gradient-thermal strain method(GTG-TSM)-achieved improved computational efficiency and consistent precision.Furthermore,the proposed method required much less computation time than the traditional TEP-FEM.Thus,this study lays the foundation for future prediction of welding deformation in more complex marine thin plates.展开更多
基于RSSI(Received Signal Strength Indication)位置指纹的Wi-Fi室内定位现已被大量应用于各类基于位置信息的服务中。但指纹定位的精度受到RSSI信号的剧烈波动影响,难以满足高精度位置信息服务的需求。为克服该困难,提出一种结合虚拟A...基于RSSI(Received Signal Strength Indication)位置指纹的Wi-Fi室内定位现已被大量应用于各类基于位置信息的服务中。但指纹定位的精度受到RSSI信号的剧烈波动影响,难以满足高精度位置信息服务的需求。为克服该困难,提出一种结合虚拟AP技术与高精度CNN(Convolutional Neural Network)判别模型的定位方法。该方法通过距离比定位得到虚拟AP的位置,并将该信息与RSSI融合作为数据增强CNN模型的输入,确定样本的位置。设计实验方案采集实际的用户终端RSSI数据,构建指纹定位的数据集,验证所提出的指纹定位方案的有效性。实验结果表明,在该数据集上,所提出的方法在确定区域时的准确度达到91%,并将95%的定位误差控制在2 m以内。对比现有的定位方案,所提出的方案在定位精度上有显著提升。展开更多
文摘Pedestrian detection has been a hot spot in computer vision over the past decades due to the wide spectrum of promising applications,and the major challenge is false positives that occur during pedestrian detection.The emergence of various Convolutional Neural Network-based detection strategies substantially enhances pedestrian detection accuracy but still does not solve this problem well.This paper deeply analyzes the detection framework of the two-stage CNN detection methods and finds out false positives in detection results are due to its training strategy misclassifying some false proposals,thus weakening the classification capability of the following subnetwork and hardly suppressing false ones.To solve this problem,this paper proposes a pedestrian-sensitive training algorithm to help two-stage CNN detection methods effectively learn to distinguish the pedestrian and non-pedestrian samples and suppress the false positives in the final detection results.The core of the proposed algorithm is to redesign the training proposal generating scheme for the two-stage CNN detection methods,which can avoid a certain number of false ones that mislead its training process.With the help of the proposed algorithm,the detection accuracy of the MetroNext,a smaller and more accurate metro passenger detector,is further improved,which further decreases false ones in its metro passenger detection results.Based on various challenging benchmark datasets,experiment results have demonstrated that the feasibility of the proposed algorithm is effective in improving pedestrian detection accuracy by removing false positives.Compared with the existing state-of-the-art detection networks,PSTNet demonstrates better overall prediction performance in accuracy,total number of parameters,and inference time;thus,it can become a practical solution for hunting pedestrians on various hardware platforms,especially for mobile and edge devices.
基金supported in part by the National Natural Science Foundation(Nos.62271248,62401256)in part by the Natural Science Foundation of Ji-angsu Province(Nos.BK20230090,BK20241384)in part by the Key Laboratory of Land Satellite Remote Sens-ing Application,Ministry of Natural Resources of China(No.KLSMNR-K202303)。
文摘In recent years,deeps learning has been widely applied in synthetic aperture radar(SAR)image processing.However,the collection of large-scale labeled SAR images is challenging and costly,and the classification accuracy is often poor when only limited SAR images are available.To address this issue,we propose a novel framework for sparse SAR target classification under few-shot cases,termed the transfer learning-based interpretable lightweight convolutional neural network(TL-IL-CNN).Additionally,we employ enhanced gradient-weighted class activation mapping(Grad-CAM)to mitigate the“black box”effect often associated with deep learning models and to explore the mechanisms by which a CNN classifies various sparse SAR targets.Initially,we apply a novel bidirectional iterative soft thresholding(BiIST)algorithm to generate sparse images of superior quality compared to those produced by traditional matched filtering(MF)techniques.Subsequently,we pretrain multiple shallow CNNs on a simulated SAR image dataset.Using the sparse SAR dataset as input for the CNNs,we assess the efficacy of transfer learning in sparse SAR target classification and suggest the integration of TL-IL-CNN to enhance the classification accuracy further.Finally,Grad-CAM is utilized to provide visual explanations for the predictions made by the classification framework.The experimental results on the MSTAR dataset reveal that the proposed TL-IL-CNN achieves nearly 90%classification accuracy with only 20%of the training data required under standard operating conditions(SOC),surpassing typical deep learning methods such as vision Transformer(ViT)in the context of small samples.Remarkably,it even presents better performance under extended operating conditions(EOC).Furthermore,the application of Grad-CAM elucidates the CNN’s differentiation process among various sparse SAR targets.The experiments indicate that the model focuses on the target and the background can differ among target classes.The study contributes to an enhanced understanding of the interpretability of such results and enables us to infer the classification outcomes for each category more accurately.
基金Supported in part by Natural Science Foundation of Guangxi(2023GXNSFAA026246)in part by the Central Government's Guide to Local Science and Technology Development Fund(GuikeZY23055044)in part by the National Natural Science Foundation of China(62363003)。
文摘In this paper,we consider the maximal positive definite solution of the nonlinear matrix equation.By using the idea of Algorithm 2.1 in ZHANG(2013),a new inversion-free method with a stepsize parameter is proposed to obtain the maximal positive definite solution of nonlinear matrix equation X+A^(*)X|^(-α)A=Q with the case 0<α≤1.Based on this method,a new iterative algorithm is developed,and its convergence proof is given.Finally,two numerical examples are provided to show the effectiveness of the proposed method.
基金Supported by the Natural Science Foundation of Guangxi Province(Grant Nos.2023GXNSFAA026067,2024GXN SFAA010521)the National Natural Science Foundation of China(Nos.12361079,12201149,12261026).
文摘Convex feasibility problems are widely used in image reconstruction,sparse signal recovery,and other areas.This paper is devoted to considering a class of convex feasibility problem arising from sparse signal recovery.We rst derive the projection formulas for a vector onto the feasible sets.The centralized circumcentered-reection method is designed to solve the convex feasibility problem.Some numerical experiments demonstrate the feasibility and e ectiveness of the proposed algorithm,showing superior performance compared to conventional alternating projection methods.
基金Supported by the National Natural Science Foundation of China under Grant No.51975138the High-Tech Ship Scientific Research Project from the Ministry of Industry and Information Technology under Grant No.CJ05N20the National Defense Basic Research Project under Grant No.JCKY2023604C006.
文摘Marine thin plates are susceptible to welding deformation owing to their low structural stiffness.Therefore,the efficient and accurate prediction of welding deformation is essential for improving welding quality.The traditional thermal elastic-plastic finite element method(TEP-FEM)can accurately predict welding deformation.However,its efficiency is low because of the complex nonlinear transient computation,making it difficult to meet the needs of rapid engineering evaluation.To address this challenge,this study proposes an efficient prediction method for welding deformation in marine thin plate butt welds.This method is based on the coupled temperature gradient-thermal strain method(TG-TSM)that integrates inherent strain theory with a shell element finite element model.The proposed method first extracts the distribution pattern and characteristic value of welding-induced inherent strain through TEP-FEM analysis.This strain is then converted into the equivalent thermal load applied to the shell element model for rapid computation.The proposed method-particularly,the gradual temperature gradient-thermal strain method(GTG-TSM)-achieved improved computational efficiency and consistent precision.Furthermore,the proposed method required much less computation time than the traditional TEP-FEM.Thus,this study lays the foundation for future prediction of welding deformation in more complex marine thin plates.
