摘要
在众多双目影像密集匹配算法中,半全局匹配(Semi-Global Matching,SGM)算法相比局部匹配算法和全局匹配算法在精度和效率方面具有较高优势。相较于一般影像,城市场景影像往往存在大量人造建筑物。针对城市区域立体影像的这一特点,本文提出一种基于Hough变换直线检测的城区影像SGM改进算法,将Hough变换获得的直线信息融入SGM的匹配框架,提高密集匹配算法的精确度和准确度。
In many binocular image dense matching algorithms,Semi-Global Matching(SGM)algorithm has higher accuracy and efficiency than local matching algorithm and global matching algorithm.Compared with general images,there are a lot of man-made buildings in urban scene images.According to the characteristics of urban stereo images,this paper proposes an improved SGM algorithm based on Hough transform line detection,which integrates the line information obtained by Hough transform into the matching framework of SGM to improve the accuracy and precision of dense matching algorithm.
作者
敖博
钱建国
谢林甫
王伟玺
李晓明
AO Bo;QIAN Jianguo;XIE Linfu;WANG Weixi;LI Xiaoming(School of Geomatics,Liaoning Technology University,Fuxin 123000,China;Institute of Smart City,School of Architecture and Urban Planning,Shenzhen University,Shenzhen 518061,China)
出处
《测绘与空间地理信息》
2023年第1期148-150,154,共4页
Geomatics & Spatial Information Technology
