摘要
运动参数估计在图像配准,电子稳像以及图像拼接中起着很重要的作用,其中块匹配算法是经常使用的一种方法。当平移较大的时候,块匹配算法仍然存在不足,因此本文提出了一种新的算法。首先利用灰度投影法得到相邻两幅图像的平移量,在此基础上再利用块匹配算法进行分块处理。实验说明了此方法的有效性,能够满足一般要求。
Motion parameter estimation play an important role in image matching, electric image stabilizing and image mosaic, and the block-matching algorithm is the most used method. In the case of large translation between two images, the block-matching algo- rithm is insufficient. In this paper, we propose a new method. First, we calculate the translation between two images by gray projection algorithm, then we use the block-matching algorithm in the divided blocks. The experiment result show that this method is efficient, and satisfy the general demand.
出处
《微计算机信息》
北大核心
2007年第02X期269-270,共2页
Control & Automation
关键词
运动参数估计
灰度投影
块匹配
motion parameter estimation,gray projection,block-matching algorithm
