电子稳像的灰度投影三点局域自适应搜索算法

Three-point locally adaptive searching in gray scale projection algorithm for electronic image stabilization

灰度投影算法是检测图像序列帧间运动矢量,实现电子稳像的关键技术之一。通过对其相关曲线的分析,基于单峰性的特征,提出了三点局域自适应搜索算法,即:在搜索宽度内均匀选取3点,计算比较其相关值,判断相关曲线梯度的变化方向,相应调整搜索参数,获得帧间运动矢量,在此基础上即可进行运动滤波和图像补偿,从而实现电子稳像。当搜索宽度为2n-1(n=1、2……且2n-1小于搜索方向上图像宽度的一半)时其运算量仅为3n,而非以往全局搜索算法的2n+1-1,从而克服了运算量大、耗时长的缺陷。该算法得到了仿真实验的验证。

The gray scale projection algorithm is a key technique for detecting motion vector of image sequential frames and realize electronic image stabilization. Through analyzing the related curves, three-point locally adaptive searching algorithm is proposed on the basis of single peak characteristics. The procedures are as follows: through equably selecting three points in searching coverage, calculating and comparing their correlation values, judging the variation direction of the correlation curve gradient, the searching parameters will be correspondingly adjusted to obtain the motion vector between frames. And then motion filtering and image compensation can be carried out on this basis and electronic image stabilization can be performed. When searching coverage is 2n-1(n=1, 2,厖,and 2n-1 less than one half of image width in searching direction), its operation load is 3n instead of 2n+1-1 of global searching algorithm. In this way the faults of large operation load and long operation time have been overcome. The algorithm has been demonstrated by simulation experiments.