计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (28): 34-36.

• 学术探讨 • 上一篇    下一篇

基于边缘拟合直线的遥感图像自动配准

杨 欣,唐金辉,牛宗标,吴秀清   

  1. 中国科学技术大学 电子工程与信息科学系,合肥 230027
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-10-01 发布日期:2007-10-01
  • 通讯作者: 杨 欣

Remote sensing image registration based on fitted straight lines of edges

YANG Xin,TANG Jin-hui,NIU Zong-biao,WU Xiu-qing   

  1. Department of Electronic Engineering & Information Science of USTC,Hefei 230027,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-10-01 Published:2007-10-01
  • Contact: YANG Xin

摘要: 旋转问题是目前图像自动配准中的一个难点,普通方法计算复杂度大,无法满足很多应用中实时性的要求,而基于边缘方向直方图的方法对噪声很敏感,对边缘进行直线拟合,提出了一种新的基于边缘拟合直线的配准算法。提出了边缘拟合直线角度差直方图的构建方法,并利用它对待配准图像进行旋转角度补偿,然后使用边缘点集的Hausdorff距离进行平移量估计,获得变换参数的近似值;最后以基于Renyi熵的互信息为相似性测度,利用遗传算法在粗配准的近似解的小范围邻域内快速寻找最优解进行精确配准。该方法计算简单、抗噪声性能强且能获得较好的配准精度。

关键词: 配准, 边缘直线拟合, 角度差直方图, Hausdorff距离

Abstract: Rotation problem is a difficulty in common image registration methods for its computational complexity,which can’t satisfy some real-time applications.In this paper,a novel method based on fitted straight lines of edges is proposed.It first compensates the rotating angle by the angle difference histogram of fitted straight lines,then estimates the translation by Hausdorff distance,finally introduces mutual information based on Renyi entropy and uses gene arithmetic to find the optimization in the small domain.The experiment results demonstrate that the new method is applicable in many applications.

Key words: registration, edge lines fitting, angle histogram, Hausdorff distance