Tracking method of moving target in complex environment
Zhuo Yan, Xiangdong Zhou, Jiahu Yuan, Xiaowei Han · 2017
Implementing robust and real-time visual tracking method is a challenging task due to many disturbing factors such as illumination changes, appearance changes, noises and irregular dynamic occlusion, etc. In order to solve this problem, a moving target tracking method, which can update the background template real-time with memory and establish self-adaptive background template library, is presented in this paper. Firstly, erosion processing, boundary tracing, sparse processing and full-matrix operating are used to determine moving target in an image. Secondly, a self-adaptive template library is established. Finally, the background template is updated at the same time. Experimental results show that the proposed method can achieve effective tracking task with real-time requirement, and can well respond to many nature interferences.