基于彩色信息尺度不变特征的街景影像匹配  

Street View Images Matching Algorithm Based on Color Scale-invariant Feature Transform

作　　者：何培培[1] 万幼川[1] 高贤君[1] 秦家鑫[1] 

机构地区：[1]武汉大学遥感信息工程学院,湖北武汉430079

出　　处：《武汉大学学报（信息科学版）》2014年第7期867-872,共6页Geomatics and Information Science of Wuhan University

基　　金：国家科技支撑计划资助项目(2012BAH34B02;2012BAJ15B04;2011BAH12B03)~~

摘　　要：提出了一种基于彩色信息的尺度不变特征匹配算法以实现彩色图像间的特征匹配。引入对光照变化具有一定鲁棒性的对角-偏移模型,通过模型变换求出图像各位置的彩色不变量,并以此作为输入图像,再采用改进的Harris算法提取图像的几何信息,建立一种具有几何和颜色不变性的特征向量,最后利用特征向量间的相关系数完成匹配。本文将该算法用于街景影像匹配中,并对比与传统SIFT算法的差异,结果表明此算法不但能够得到分布均匀密集度高的匹配点,且耗时少,匹配精度高,证明了对于街景影像匹配研究的适用性。Mage feature extraction and matching will be done after color image is transformed into gray may cause color information loss and lead to wrong matching.In order to solve problem,this paper presents a method color-based scale-Invariant feature transform to realize feature image matching.Then,feature points are extracted by using improved Harris algorithm eighbor information around these points is described using color invariant value as input information.Finally,the points between two images are matched using correlation coefficient.To demonstrate the feasibility of approach,experiments with groups of street view imagescompared with the traditional SIFT algorithm.show that the proposed algorithm can not only intensive match points,higher matching correct rate speed validity.

关 键 词：彩色不变量 尺度不变性 特征提取 图像匹配 街景影像 

分 类 号：P237.3[天文地球—摄影测量与遥感] P234.1[天文地球—测绘科学与技术]