基于改进的ORB算法的红外遥感图像拼接研究  

Research on Infrared Remote Sensing Image Mosaicing Based on Improved ORB Algorithm

作　　者：彭畅 李广泽[1] 张晓阳 左洋[1] PENG Chang;LI Guang-ze;ZHANG Xiao-yang;ZUO Yang(College of Daheng,Changchun Institute of Optics Fine Mechanics and Physics Chinese Academy of Sciences,Changchun 130033,China;College of Materials,University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院长春光学精密机械与物理研究所大珩学院,吉林长春130033 [2]中国科学院大学材料学院,北京100049

出　　处：《控制工程》2020年第8期1332-1336,共5页Control Engineering of China

基　　金：国家自然科学基金资助项目(61405191)。

摘　　要：基于红外遥感图像提出一种快速、高效的拼接算法,针对SIFT、SURF算法运行时间久的缺点,提出基于ORB特征点匹配算法,对红外遥感图像特征点对进行匹配,接着采用RANSAC算法对提取出的特征点进行误匹配剔除,最终采用渐入渐出的加权融合算法实现了红外遥感图像拼接。该算法在保证图像拼接质量的基础上,提高了算法的运行速度,可以实现快速、高效的红外遥感图像拼接。A fast and efficient efficient algorithm based on infrared remote sensing image is presented in this paper.Aiming at SIFT and SURF algorithm running for a long time,this paper proposes an ORB feature point matching algorithm,which matches feature points of infrared remote sensing images,then the RANSAC algorithm is used to delete the extracted feature points.Incremental weighting fusion algorithm is finally used to achieve infrared remote sensing image mosaic.The algorithm improves the running speed of the original algorithm on the basis of ensuring the image mosaic quality,and can achieve fast and efficient infrared remote sensing image splicing.

关 键 词：红外遥感图像 ORB RANSAC 渐入渐出加权 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]