野外科研视频中鸟类目标的检测和跟踪技术研究

Wild Birds Target Detection and Tracking Technology in Video System

野外科研视频监控系统能帮助科研人员更好地进行长期的考察研究工作,但观看研究海量的视频数据也是很费时费力的重复性工作,为此,我们研究了科研视频中运动目标的检测和跟踪技术,使用背景剪除法和连续适应性均值移动算法对青海湖鸟类监控视频进行了斑头雁目标的检测和跟踪,经过实际监控视频的验证,该方法可以较好地检测和跟踪斑头雁目标,为后续的鸟类行为研究工作打下了基础。

Through the establishment of video sensor networks, it can allow researchers to do long-term field observations. But the study of mass video data is very costly and time-consuming repetitive work. For this, we studied the moving object detection and tracking in research video, using background subtraction method and continuous adaptive mean shift algorithm for target detection and tracking of the bar-headed goose in the Qinghai Lake bird surveillance video. Veriifcation of the actual video shows that the method can be effectively used to detect and track the target of Bar-headed goose, this would be a good base of future study of wild birds behavior.