中文摘要：

RANSAC是一种鲁棒性估计算法,常用于可见光图像的匹配。文章将其用在红外图像的匹配过程中,并根据红外图像清晰度差,纹理信息少等特点,改进了该算法,提出了一种分区域的RANSAC算法。应用HARRIS算子提取特征点,在匹配过程中将人脸划分为不同区域,应用RANSAC算法进行匹配。实验仿真结果表明,此算法在红外热图像的匹配上,具有准确率高,计算量小的优点,有红外热图像的建模上有较高使用价值。

英文摘要：

RANSAC is a robustness estimate algorithm,always used in visible light images matching.In this paper,it is used in infrared images matching,and as the infrared images have low definition and lack of texture information,a subarea RANSAC matching algorithm is presented.In this algorithm,Harris algorithm is used in feature extraction.Then the aim face is divided into several areas,feature matching by RANSAC algorithm.Experimental results show this algorithm has high correct rates and low amount of calculation in matching of infrared images,and has high valuable in infrared modeling.