摘要
针对视频运动目标的跟踪问题,提出了一种基于改进Hausdorff距离的运动目标自动跟踪算法。首先对视频图像作预处理,使用Canny算子进行边缘提取;然后采用对称差分算法建立运动目标的初始模板;接着采用改进的Hausdorff距离和新的匹配搜索策略,实现后续帧中运动目标与初始模板的匹配;最后,采用一种新的模板更新算法———4步法对模板进行更新。实验结果表明,该算法不仅能准确跟踪运动目标,而且能有效提高目标跟踪效率。
In order to solve the video moving object tracking problem,a method based on improved Hausdorff distance moving object auto tracking algorithm is presented. First, preprocess the video by using Canny operator edge extraction. After establishing the initial template of moving obiects by symmetric difference algorithm, we employ the improved Hausdorff distance and the new matching search strategy to match the moving object in subsequent frames with initial template. Finally, four steps algorithm,a new template updating algorithm, is used to update the template. Experimental results show that the algorithm can not only accurately track moving objects, but also effectively improve the tracking efficiency.
出处
《电子测量技术》
2014年第4期56-60,共5页
Electronic Measurement Technology
