An Effective Target Tracking Algorithm in Infrared Images Video
Yuhui Liu, Qianqian Yan, Wei Liu, Huai Yuan, Guangyuan Zhang · 2010
In order to track a target in a complex background, an effective target tracking algorithm in infrared images video is presented in this paper. A new similarity measurement algorithm based on normalized cross-correlation and distance was proposed, and the tracking confidence was designed and the adaptive threshold adjustment strategy of template updating was obtained. Also, the adaptive template updating algorithm which was based on the confidence of tracking was achieved. In addition, Kalman filtering is used to predict the position of the target, in order to resolve the temporary occlusion problem of the moving target, ensuring continuous tracking. Experimental results showed that the algorithm can adapt to such situations as target deformation or being blocked and so on, achieving an effective and robust target tracking.