红外技术, 2011, 33 (5): 271, 网络出版: 2011-06-24
基于改进配准测度的红外与可见光图像配准
A Registration Algorithm of Infrared and Visible Images Based on Revised Measure Function
特征点提取 配准测度 特征点匹配 配准参数 gain feature points registration measure matching feature point pairs registration parameters
摘要
针对在可见光与红外配准中采用传统测度算法时存在配准精度低的问题, 提出了改进配准测度函数(RMF), 实现可见光和红外图像高精度自适应配准。首先选用harris 算子提取两幅图像的特征点;再利用改进的配准测度函数(RMF)有效的获取特征点对;最后利用RMF 最大的两对匹配点计算配准参数, 并进行图像配准。实验结果证明了该方法的实用性, 特别对于任意旋转角、缩放比例和偏移量的图像配准。
Abstract
Aiming at the problem of low accuracy caused by traditional measure functions in infrared and visible images registration, we proposed a new Revised Measure Function(RMF), which is an accurate and automated image registration method for infrared and visible images. We first select harris operator to detect feature points in two images, then use revised measure function to gain matching feature point pairs, and finally calculate registration parameters at two register point pairs with maximum RMF value. Experiment results show its validity and accuracy, especially for register images with arbitrary rotation angle, zoom size and translation value.
韩静, 柏连发, 张毅, 陈钱. 基于改进配准测度的红外与可见光图像配准[J]. 红外技术, 2011, 33(5): 271. HAN Jing, BO Lian-fa, ZHANG Yi, CHEN Qian. A Registration Algorithm of Infrared and Visible Images Based on Revised Measure Function[J]. Infrared Technology, 2011, 33(5): 271.