地理空间信息2025，Vol.23Issue(2)：7-10,4.DOI:10.3969/j.issn.1672-4623.2025.02.002
一种基于边缘约束的密集匹配优化方法
Dense Matching Optimization Method Based on Edge Constraints
摘要
Abstract
Dense matching is an important step in 3D reconstruction.In this study,we proposed a dense matching optimization method based on edge constraints.We used the semi-global matching(SGM)algorithm to realize dense matching at first.Then,we used the left-right consistency checking process to identify mismatched pixels,used the hybrid level set method to segment stereo pair images and generated vector polygons with relatively uniform disparity distribution.Finally,we set the search window with the radius increasing gradually with the mismatched pixel as the center,and searched the nearest correct matching point within the intersection range of search window and vector spot where the mismatched point was located.If the disparity is successfully reassigned by the nearest valid pixel found,the searching process is terminated.Comparative results obtained using well-known computer vision benchmark datasets and UAV aerial image datasets demonstrate the algorithm has high accura-cy and good robustness.关键词
航空影像/三维重建/密集匹配/影像分割/最近点搜索Key words
aerial image/3D reconstruction/dense matching/image segmentation/nearest point search分类
天文与地球科学引用本文复制引用
吴春,段春燕,胡文焯,张旭,龚睿,李璐含..一种基于边缘约束的密集匹配优化方法[J].地理空间信息,2025,23(2):7-10,4.基金项目
湖北省教育厅科研计划项目(B2019328) （B2019328）
湖北省自然资源科技项目(ZRZY2023KJ25). （ZRZY2023KJ25）
