投影相似变换的无人机影像拼接  

UAV image mosaic based on projection-similarity transformation

作　　者：王忠美[1,2,3] 杨晓梅[3] 顾行发[2] 周成虎[3] 

机构地区：[1]电子科技大学自动化工程学院,成都611731 [2]中国科学院遥感与数字地球研究所,北京100101 [3]中国科学院地理科学与资源研究所,北京100101

出　　处：《测绘科学》2016年第9期112-117,共6页Science of Surveying and Mapping

基　　金：国家"863"计划项目(2012AA121201;2013AA122901);国家科技支撑计划项目(2014BAL01B01)

摘　　要：针对当前无人机遥感影像存在的成像范围小、视差角度大的缺点,该文提出一种改进的投影-相似变换方法来进行无人机遥感图像的拼接:通过对初始获得的单应矩阵进行改进,平滑推导重叠区域的投影变换到非重叠区域形成相似变换;该变换同时考虑拼接过程中图像重叠区域的配准精度和非重叠区域的视场一致。实验结果表明,该方法的拼接性能优于传统拼接方法。Aiming at the problem of small imaging coverage and big parallax existed in current UAV image processing, this paper proposed an improved projection-similarity transformation method for UAV image mosaic： the initial homographic matrix was improved, and the projection transformation of the overlapping regions was smoothly extrapolated into the non-overlapping regions to form the similarity transformation; it considered registration precision in the overlapping area and the field of view in the non-overlapping regions together. Experimental result showed that the mosaic performance of this algorithm would be better than that of traditional algorithm.

关 键 词：无人机遥感 图像拼接 投影变换 相似变换 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]