稀疏活动轮廓扩展形状脚本模型目标检测算法

摘要

传统的稀疏活动轮廓模型可以较好地解决目标微小形变情况下的定位问题,但是对训练样本要求比较严格,且在目标发生较大形变情况下采用学习到的可变形模板对目标进行定位会产生一定偏差.针对该问题,提出一种稀疏活动轮廓扩展形状脚本模型的目标检测算法.利用勾画样本通过扩展活动轮廓模型学习到组成目标的可变形形状图案,这些形状图案构成的形状脚本模型能够清晰地定义目标模式;采用递归sum-max maps结构进行搜索,用形状脚本模型匹配测试图像实现目标定位.经过多组实验,结果表明所提算法能较好地解决目标在发生较大形变、存在遮挡以及复杂背景下的定位问题.%The traditional sparse active contour model can be used to solve the problem of localization when the target is viewed from different angle or exists a little deformation, but it is difficult to solve the problem when the target is large changed. This paper presents an object detection algorithm based on sparse active contour spread shape script model to solve this problem. Firstly, various shape motifs are learned with sketch samples by the spread active contour model, the shape script model of object is made up of the shape motifs, and so the mode of object is clearly defined. Secondly, the matching of a shape script template to a testing image can be accomplished by a cortex-like structure of recursive sum-max maps. The experimental results show that the method can solve the problem of localization when the target is large changed, occlusive or in complex background.