Abnormal event detection aims to automatically identify unusual events that do not comply with expectation.Recently,many methods have been proposed to obtain the temporal locations of abnormal events under various det...Abnormal event detection aims to automatically identify unusual events that do not comply with expectation.Recently,many methods have been proposed to obtain the temporal locations of abnormal events under various determined thresholds.However,the specific categories of abnormal events are mostly neglect,which are important to help in monitoring agents to make decisions.In this study,a Temporal Attention Network(TANet)is proposed to capture both the specific categories and temporal locations of abnormal events in a weakly supervised manner.The TANet learns the anomaly score and specific category for each video segment with only video-level abnormal event labels.An event recognition module is exploited to predict the event scores for each video segment while a temporal attention module is proposed to learn a temporal attention value.Finally,to learn anomaly scores and specific categories,three constraints are considered:event category constraint,event separation constraint and temporal smoothness constraint.Experiments on the University of Central Florida Crime dataset demonstrate the effectiveness of the proposed method.展开更多
For modern stealth aircraft,it is important to analyze the influence of Radar Cross Section(RCS)peak exposure on penetration for guiding stealth design and penetration trajectory planning,which needs to reflect the RC...For modern stealth aircraft,it is important to analyze the influence of Radar Cross Section(RCS)peak exposure on penetration for guiding stealth design and penetration trajectory planning,which needs to reflect the RCS statistical uncertainty and the RCS difference with the change of incident angle.Based on the RCS characteristics of typical stealth aircraft,this paper established a simplified RCS dynamic fluctuation statistical model with the parameters log mean and log standard deviation.According to the detection probability algorithm in radar signal processing field,this paper built the algorithm of radar detection probability based on the RCS dynamic fluctuation statistical model.The analysis of examples concluded that the key to successful penetration is to shorten the RCS peak exposure time,which can be reduced by decreasing the RCS peak width or increasing velocity.Based on the conclusion,this paper proposed the method of turning maneuvering to reduce RCS peak exposure time dramatically.展开更多
Water stable isotopes(δ^(2) H andδ^(18)O)can record surface water evaporation,which is an important hydrological process for understanding watershed structure and function evolution.However,the isotopic estimation o...Water stable isotopes(δ^(2) H andδ^(18)O)can record surface water evaporation,which is an important hydrological process for understanding watershed structure and function evolution.However,the isotopic estimation of water evaporation losses in the mountain watersheds remains poorly explored,which hinders understanding spatial variations of hydrological processes and their relationships with the temperature and vegetation.Here we investigatedδ^(2) H,δ^(18)O,and d-excess values of stream water along an altitude gradient of 2130 to 3380 m in Guan’egou mountain watershed at the east edge of the Qinghai-Tibet Plateau in China.The meanδ^(2) H(-69.6‰±2.6‰),δ^(18)O(-10.7‰±0.3‰),and dexcess values(16.0‰±1.4‰)of stream water indicate the inland moisture as the major source of precipitation in study area.Water stable isotopes increase linearly with decreasing altitudes,based on which we estimated the fractions of water evaporation losses along with the altitude and their variations in different vegetations.This study provides an isotopic evaluation method of water evaporation status in mountain watersheds,the results are useful for further understanding the relationship between hydrological processes and ecosystem function under the changing climate surrounding the Qinghai-Tibet Plateau.展开更多
基金supported in part by the National Science Fund for Distinguished Young Scholars under grant no.61925112,in part by the National Natural Science Foundation of China under grant no.61806193 and grant no.61772510Support Program of Shaanxi under grant no.2020KJXX‐091in part by the Key Research Program of Frontier Sciences,Chinese Academy of Sciences under grant no.QYZDY‐SSW‐JSC044.
文摘Abnormal event detection aims to automatically identify unusual events that do not comply with expectation.Recently,many methods have been proposed to obtain the temporal locations of abnormal events under various determined thresholds.However,the specific categories of abnormal events are mostly neglect,which are important to help in monitoring agents to make decisions.In this study,a Temporal Attention Network(TANet)is proposed to capture both the specific categories and temporal locations of abnormal events in a weakly supervised manner.The TANet learns the anomaly score and specific category for each video segment with only video-level abnormal event labels.An event recognition module is exploited to predict the event scores for each video segment while a temporal attention module is proposed to learn a temporal attention value.Finally,to learn anomaly scores and specific categories,three constraints are considered:event category constraint,event separation constraint and temporal smoothness constraint.Experiments on the University of Central Florida Crime dataset demonstrate the effectiveness of the proposed method.
文摘For modern stealth aircraft,it is important to analyze the influence of Radar Cross Section(RCS)peak exposure on penetration for guiding stealth design and penetration trajectory planning,which needs to reflect the RCS statistical uncertainty and the RCS difference with the change of incident angle.Based on the RCS characteristics of typical stealth aircraft,this paper established a simplified RCS dynamic fluctuation statistical model with the parameters log mean and log standard deviation.According to the detection probability algorithm in radar signal processing field,this paper built the algorithm of radar detection probability based on the RCS dynamic fluctuation statistical model.The analysis of examples concluded that the key to successful penetration is to shorten the RCS peak exposure time,which can be reduced by decreasing the RCS peak width or increasing velocity.Based on the conclusion,this paper proposed the method of turning maneuvering to reduce RCS peak exposure time dramatically.
基金support by Tanchang County People’s Government,Forestry Bureau of Tanchang County,and Guan’egou National Forest Park on the field worksupported by National Natural Science Foundation of China(No.41730855)State Key Project of Research and Development Plan(2016YFA0600802)。
文摘Water stable isotopes(δ^(2) H andδ^(18)O)can record surface water evaporation,which is an important hydrological process for understanding watershed structure and function evolution.However,the isotopic estimation of water evaporation losses in the mountain watersheds remains poorly explored,which hinders understanding spatial variations of hydrological processes and their relationships with the temperature and vegetation.Here we investigatedδ^(2) H,δ^(18)O,and d-excess values of stream water along an altitude gradient of 2130 to 3380 m in Guan’egou mountain watershed at the east edge of the Qinghai-Tibet Plateau in China.The meanδ^(2) H(-69.6‰±2.6‰),δ^(18)O(-10.7‰±0.3‰),and dexcess values(16.0‰±1.4‰)of stream water indicate the inland moisture as the major source of precipitation in study area.Water stable isotopes increase linearly with decreasing altitudes,based on which we estimated the fractions of water evaporation losses along with the altitude and their variations in different vegetations.This study provides an isotopic evaluation method of water evaporation status in mountain watersheds,the results are useful for further understanding the relationship between hydrological processes and ecosystem function under the changing climate surrounding the Qinghai-Tibet Plateau.
基金supported by the Science and Technology Major Project of Fujian Province,China (2022HZ027006)Fujian Provincial Science and Technology Planning Project (2022I0006)+1 种基金Quanzhou Municipal Science and Technology Major Project,China (2022GZ7)the National Natural Science Foundation of China (62274036)。