基于三参数模型的视频图像配准方法研究

Image registration based on 3-parameter motion model

视频图像配准是运动视频处理中的一项关键技术。提出了一种新的基于3参数模型的配准算法,该算法利用多尺度的角点检测方法从相邻两帧中抽取特征点,选用Hausdorff距离对特征点集进行匹配,计算3参数模型参数,从而实现图像配准。与传统的6参数仿射模型相比,模型参数的降低并没有显著降低配准效果,同时由于参数搜索空间的减少,该方法用更快的速度得到较准确的结果。

Image registration is a key technique for some video processing applications.This paper presents a new method based on 3-parameter. motion model.First,feature points are extracted in both images using Harris corner based on the theory of scalespace.Then,these feature points are matched based on the Hausdorff distance,and the 3-parameter motion model is obtained. Compared with the affine transformation motion model,decreasing of parameters doesn＇t affect the effect of image registration.On the same time ,the calculation speed of image registration is improved as the decreasing of parameters.