火力与指挥控制2023，Vol.48Issue(12)：79-84,6.DOI:10.3969/j.issn.1002-0640.2023.12.012
基于航拍视频的车辆目标跟踪方法
A Vehicle Target Tracking Method Based on Aerial Video
摘要
Abstract
Taking the moving vehicle photographed by UAV as the research object and aiming at a certain vehicle target in aerial video,the target tracking method is studied.Firstly,in order to accurately predict the position of the vehicle in the future frame,a target prediction algorithm based on K-Neighborhood is presented.Secondly,for the situation that the vehicle target may be completely blocked,an anti occlusion algorithm based on vehicle motion state estimation is proposed.Finally,in the process of target detection,aiming at the situation that multiple targets may be detected,an image matching algorithm based on multi feature fusion is given to match the detection results.The above algorithms are combined,a vehicle target tracking method is formed.The test results show that the average accuracy of the target tracking method proposed in this paper reaches 91.1%,which is about 6.35%higher than that of the common target tracking algorithms.关键词
目标检测/目标跟踪/图像匹配/抗遮挡Key words
target detection/target tracking/image matching/anti occlusion分类
信息技术与安全科学引用本文复制引用
吕艳辉,贾旭东,李彬..基于航拍视频的车辆目标跟踪方法[J].火力与指挥控制,2023,48(12):79-84,6.基金项目
辽宁省教育厅基金资助项目(LJKZ0241) （LJKZ0241）
