Multi-object tracking(MOT) techniques have been increasingly applied in a diverse range of tasks. Unmanned aerial vehicle(UAV) is one of its typical application scenarios. Due to the scene complexity and the low resol...Multi-object tracking(MOT) techniques have been increasingly applied in a diverse range of tasks. Unmanned aerial vehicle(UAV) is one of its typical application scenarios. Due to the scene complexity and the low resolution of moving targets in UAV applications, it is difficult to extract target features and identify them. In order to solve this problem, we propose a new re-identification(re-ID) network to extract association features for tracking in the association stage. Moreover, in order to reduce the complexity of detection model, we perform the lightweight optimization for it. Experimental results show that the proposed re-ID network can effectively reduce the number of identity switches, and surpass current state-of-the-art algorithms. In the meantime, the optimized detector can increase the speed by 27% owing to its lightweight design, which enables it to further meet the requirements of UAV tracking tasks.展开更多
A distinct aridity trend in China in last 100 years is presented by applying a linear fitting to both the climate records and the hydrological records, which is supported by evidence of environmental changes and seems...A distinct aridity trend in China in last 100 years is presented by applying a linear fitting to both the climate records and the hydrological records, which is supported by evidence of environmental changes and seems to be associated with a global warming trend during this period.The Mann Kendall Rank statistic test reveals a very interesting feature that the climate of China entered into a dry regime abruptly in about 1920's, which synchronized with the rapid warming of the global temperature at almost the same time.According to an analysis of the meridional profile of observed global zonal mean precipitation anomalies during the peak period of global wanning (1930-1940), the drought occurred in whole middle latitude zone (25°N-55°N) of the Northern Hemisphere, where the most part of China is located in. Although this pattern is in good agreement with the latitude distribution of the difference of zonal mean rates of precipitation between 4 × CO2 and 1 × CO2 simulated by climate model (Manabe and Wetherald, 1983), more studies are required to understand the linkage between the aridity trend in China and the greenhouse effect.The EOF analysis of the Northern Hemisphere sea level pressure for the season of June to August shows an abrupt change of the time coefficient of its first eigenvector from positive to negative in mid-1920's, indicating an enhancement of the subtropical high over Southeast Asia and the western Pacific after that time. This is an atmospheric circulation pattern that is favorable to the development of dry climate in China.展开更多
We are very pIeased to announce that five Associate Editors have joined the editorial team of the Joumal of Environmental Sciences （JES）. Professors/Drs. Yong Cai, Paul Lain, Jonathan Martin, Michael Plewa, and Po K...We are very pIeased to announce that five Associate Editors have joined the editorial team of the Joumal of Environmental Sciences （JES）. Professors/Drs. Yong Cai, Paul Lain, Jonathan Martin, Michael Plewa, and Po Keung Wong bring a wealth of expertise in environmental sciences. As preeminent scientists in their chosen areas of research.展开更多
基金supported by the Research Foundation of Nanjing University of Posts and Telecommunications (No.NY219076)。
文摘Multi-object tracking(MOT) techniques have been increasingly applied in a diverse range of tasks. Unmanned aerial vehicle(UAV) is one of its typical application scenarios. Due to the scene complexity and the low resolution of moving targets in UAV applications, it is difficult to extract target features and identify them. In order to solve this problem, we propose a new re-identification(re-ID) network to extract association features for tracking in the association stage. Moreover, in order to reduce the complexity of detection model, we perform the lightweight optimization for it. Experimental results show that the proposed re-ID network can effectively reduce the number of identity switches, and surpass current state-of-the-art algorithms. In the meantime, the optimized detector can increase the speed by 27% owing to its lightweight design, which enables it to further meet the requirements of UAV tracking tasks.
文摘A distinct aridity trend in China in last 100 years is presented by applying a linear fitting to both the climate records and the hydrological records, which is supported by evidence of environmental changes and seems to be associated with a global warming trend during this period.The Mann Kendall Rank statistic test reveals a very interesting feature that the climate of China entered into a dry regime abruptly in about 1920's, which synchronized with the rapid warming of the global temperature at almost the same time.According to an analysis of the meridional profile of observed global zonal mean precipitation anomalies during the peak period of global wanning (1930-1940), the drought occurred in whole middle latitude zone (25°N-55°N) of the Northern Hemisphere, where the most part of China is located in. Although this pattern is in good agreement with the latitude distribution of the difference of zonal mean rates of precipitation between 4 × CO2 and 1 × CO2 simulated by climate model (Manabe and Wetherald, 1983), more studies are required to understand the linkage between the aridity trend in China and the greenhouse effect.The EOF analysis of the Northern Hemisphere sea level pressure for the season of June to August shows an abrupt change of the time coefficient of its first eigenvector from positive to negative in mid-1920's, indicating an enhancement of the subtropical high over Southeast Asia and the western Pacific after that time. This is an atmospheric circulation pattern that is favorable to the development of dry climate in China.
文摘We are very pIeased to announce that five Associate Editors have joined the editorial team of the Joumal of Environmental Sciences （JES）. Professors/Drs. Yong Cai, Paul Lain, Jonathan Martin, Michael Plewa, and Po Keung Wong bring a wealth of expertise in environmental sciences. As preeminent scientists in their chosen areas of research.
