基于Hausdorff距离的非刚体目标自适应轮廓跟踪

本文提出了一种基于Ｈａｕｓｄｏｒｆ距离的非刚体目标的轮廓跟踪算法。它的特点在于从二维序列图像中提取非刚体目标的二维可变模板，从而实现对非刚体运动目标的跟踪。此算法的主要思想在于将空间运动的非刚体目标的二维图像变化分解成二维运动变化和二维形状变化，从而加大了模板适应力，辅以过零点检测及金字塔快速搜索算法，可以实现非刚体目标的快速跟踪。最后，本文给出了对人体轮廓进行跟踪的实验结果。

Adaptive Tracking of Non-Rigid Objects Contour based on Hausdorff Distance

A contour tracking method based on Hausdorff distance to track non rigid objects moving in the two dimensional images is described.The main idea of this method is to decompose the variation of the non rigid object contour in space into two dimensional shape changes and two dimensional motions.So the templates adaptability are improved. The fast tracking can be obtained combined by the pyramid algorithm and zero crossing detection algorithm.The experiments conducted on human contour tracking are reported.