Deep Learning(DL)model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR)and has achieved excellent performance.However,the black-box nature of DL models has been the f...Deep Learning(DL)model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR)and has achieved excellent performance.However,the black-box nature of DL models has been the focus of criticism,especially in the application of SARATR,which is closely associated with the national defense and security domain.To address these issues,a new interpretable recognition model Physics-Guided BagNet(PGBN)is proposed in this article.The model adopts an interpretable convolutional neural network framework and uses time–frequency analysis to extract physical scattering features in SAR images.Based on the physical scattering features,an unsupervised segmentation method is proposed to distinguish targets from the background in SAR images.On the basis of the segmentation result,a structure is designed,which constrains the model's spatial attention to focus more on the targets themselves rather than the background,thereby making the model's decision-making more in line with physical principles.In contrast to previous interpretable research methods,this model combines interpretable structure with physical interpretability,further reducing the model's risk of error recognition.Experiments on the MSTAR dataset verify that the PGBN model exhibits excellent interpretability and recognition performance,and comparative experiments with heatmaps indicate that the physical feature guidance module presented in this article can constrain the model to focus more on the target itself rather than the background.展开更多
By analyzing the bus operation environment and accounting for prediction uncertainties,a bus arrival interval prediction model was developed utilizing a gated recur-rent unit(GRU)neural network.To reduce the impact of...By analyzing the bus operation environment and accounting for prediction uncertainties,a bus arrival interval prediction model was developed utilizing a gated recur-rent unit(GRU)neural network.To reduce the impact of irrelevant data and boost prediction accuracy,an attention mechanism was integrated into the point model to concen-trate on important input sequence information.Based on the point predictions,the lower upper bound estimation(LUBE)method was used,providing a range for the bus interval times predicted by the model.The model was vali-dated using data from 169 bus routes in Nanchang,Jiangxi Province.The results indicated that the attention-GRU model outperformed neural network,long short-term memory and GRU models.Compared with the Bootstrap method,the LUBE method has a narrower average interval width.The coverage width-based criterion(CWC)was reduced by 8.1%,2.2%,and 5.7%at confidence levels of 85%,90%,and 95%,respectively,during the off-peak period,and by 23.2%,26.9%,and 27.3%at confidence levels of 85%,90%,and 95%,respectively,during the peak period.Therefore,it can accurately describe the fluctuation range in bus arrival times with higher accuracy and stability.展开更多
基金co-supported by the National Natural Science Foundation of China(No.62001507)the Youth Talent Lifting Project of the China Association for Science and Technology(No.2021-JCJQ-QT-018)+1 种基金the Program of the Youth Innovation Team of Shaanxi Universitiesthe Natural Science Basic Research Plan in Shaanxi Province of China(No.2023-JC-YB-491)。
文摘Deep Learning(DL)model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR)and has achieved excellent performance.However,the black-box nature of DL models has been the focus of criticism,especially in the application of SARATR,which is closely associated with the national defense and security domain.To address these issues,a new interpretable recognition model Physics-Guided BagNet(PGBN)is proposed in this article.The model adopts an interpretable convolutional neural network framework and uses time–frequency analysis to extract physical scattering features in SAR images.Based on the physical scattering features,an unsupervised segmentation method is proposed to distinguish targets from the background in SAR images.On the basis of the segmentation result,a structure is designed,which constrains the model's spatial attention to focus more on the targets themselves rather than the background,thereby making the model's decision-making more in line with physical principles.In contrast to previous interpretable research methods,this model combines interpretable structure with physical interpretability,further reducing the model's risk of error recognition.Experiments on the MSTAR dataset verify that the PGBN model exhibits excellent interpretability and recognition performance,and comparative experiments with heatmaps indicate that the physical feature guidance module presented in this article can constrain the model to focus more on the target itself rather than the background.
基金The National Natural Science Foundation of China(No.52162042)General Science and Technology Project of Jiangxi Provincial Department of Transportation(No.2024YB039).
文摘By analyzing the bus operation environment and accounting for prediction uncertainties,a bus arrival interval prediction model was developed utilizing a gated recur-rent unit(GRU)neural network.To reduce the impact of irrelevant data and boost prediction accuracy,an attention mechanism was integrated into the point model to concen-trate on important input sequence information.Based on the point predictions,the lower upper bound estimation(LUBE)method was used,providing a range for the bus interval times predicted by the model.The model was vali-dated using data from 169 bus routes in Nanchang,Jiangxi Province.The results indicated that the attention-GRU model outperformed neural network,long short-term memory and GRU models.Compared with the Bootstrap method,the LUBE method has a narrower average interval width.The coverage width-based criterion(CWC)was reduced by 8.1%,2.2%,and 5.7%at confidence levels of 85%,90%,and 95%,respectively,during the off-peak period,and by 23.2%,26.9%,and 27.3%at confidence levels of 85%,90%,and 95%,respectively,during the peak period.Therefore,it can accurately describe the fluctuation range in bus arrival times with higher accuracy and stability.
文摘[目的]分析新疆维吾尔自治区1986—2025年的盐碱地治理政策演变特征,以期为该区后续盐碱地治理利用工作提供科学指导。[方法]基于政策内容演变与关键节点判别法,对1986年至今公开发布的160份新疆盐碱地治理政策进行阶段划分,通过ROST Content Mining System V6.0软件的词频分析和社会语义网络分析工具,对各阶段政策进行文本挖掘与量化分析。[结果]①新疆盐碱地治理政策历经起步规划(1986—2013年)、专项整治(2014—2021年)和综合利用(2022—2025年)3阶段。②各阶段特征鲜明,起步规划阶段以土壤改良和系统性规划为核心;专项整治阶段注重耕地质量提升与生态环境保护;综合利用阶段聚焦盐碱地综合利用与高标准农田建设。③随着时间的推移,政策向全面化、精细化方向发展,治理目标转向盐碱地多元利用与可持续发展,治理过程注重与生态环境保护的平衡协调,治理框架日趋完善,治理措施逐渐多样。[结论]基于政策内容的阶段划分与文本挖掘方法可以系统揭示新疆盐碱地治理政策的发展与演变特征,具有良好的适用性与可操作性。
