主动形状模型中的一种基于概率主成分分析的对应点搜索方法

摘要

主动形状模型是一种以迭代的方式形变以匹配新图像中的目标实例的统计形状模型,它的关键是如何搜索形状模型标记点的对应点.本文提出一种新颖的采用概率主成分分析的主动形状模型的对应点搜索方法.该方法是一种基于灰度图像的对应点搜索算法,不需要对原图像进行边缘提取.在江面船舶目标的主动形状跟踪实验中,采用该方法能较准确地搜索到对应点.%Active shape models (ASMs)are statistical models of the shapes of objects which iteratively deform to fit to an example of the object in a new image. The key of the ASMs is how to search the corresponding points of the landmarks in the ASMs. In this paper,a novel method based on Probabilistic Principal Component Analysis(PPCA)for the search of the corresponding points in the ASMs is presented. This approach is a kind of gray image-based method without the need for extracting the edge. Experimental results show that the corresponding points in the ASMs for river ships tracking can be searched from image correctly using the proposed algorithm.