摘要
基于模板匹配的跟踪算法具有提取信息全面、算法简捷的优点,但目前采用固定模板匹配的算法在实际跟踪过程中容易发生误判,作者经过研究提出了自适应模板匹配的算法,该算法能有效的克服上述的问题,在此基础上作者给出了初始模板的捕获和目标物位置预测的具体算法,采用了基于自适应模板跟踪算法的“DX—1”系统在实际跟踪过程中取得了很好的跟踪效果.
The tracking algorithm based on template matching has the advantages of simpleness and obtaining more information,but now the tracking algorithm based on fix template was prone to discrimination. After researching, the author put forward a new template self-adjusting algorithm. This method can overcome the problem of the discrimination effectively. Besides, the author presents the methods of the initial model capturing and the target position forecasting. With this new algorithm, the video tracking system“DX-1” has achived good tracking effect.
出处
《重庆大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2005年第6期74-76,共3页
Journal of Chongqing University
关键词
最大化背景预测
质心
波门预测
自适应模板
匹配
maximize background prediction
center of mass
gate prediction
adaptive template
matching
