基于主颜色和纹理特征的TS模型匹配方法

摘要

In order to overcome the shortcomings of video matching methods, make up for the inaccuracy of video matching based on motion features and improve the large computing complexity of video matching based on content, and so on, this paper will put forward a TS method based on dominant color and texture feature. It will improve the efficiency of video matching, from the perspective of reducing time complexity and increasing the registration rate. By many experiments, this method solves the above problems effectively. Registration rate increases by about 20% than traditional methods, and the time complexity of algorithm reduces by about 45%.%为了弥补视频匹配方法的不足,解决基于运动特征视频匹配的不精确以及基于内容视频匹配的运算复杂度大等问题,提出一种基于主颜色和纹理特征的TS模型匹配方法,在减少时间复杂度和增加配准率基础进一步提高视频匹配的效率.大量实验证明,该方法有效的解决了上述问题,比传统方法相配准率提高约20％,算法时间复杂度降低约45％.