A fast,accurate and dense feature matching algorithm for aerial images

• 来源期刊: 系统工程与电子技术（英文版）
• 年，卷（期）: 2020,（6）
• 所属期刊栏目: ELECTRONICS TECHNOLOGY
• 页码范围: 1128-1139
• 页数: 12页
• 语种: 英文
• DOI: 10.23919/JSEE.2020.000085

摘要

Three-dimensional (3D) reconstruction based on aeri-al images has broad prospects,and feature matching is an im-portant step of it.However,for high-resolution aerial images,there are usually problems such as long time,mismatching and sparse feature pairs using traditional algorithms.Therefore,an algorithm is proposed to realize fast,accurate and dense fea-ture matching.The algorithm consists of four steps.Firstly,we achieve a balance between the feature matching time and the number of matching pairs by appropriately reducing the image resolution.Secondly,to realize further screening of the mis-matches,a feature screening algorithm based on similarity judg-ment or local optimization is proposed.Thirdly,to make the al-gorithm more widely applicable,we combine the results of differ-ent algorithms to get dense results.Finally,all matching feature pairs in the low-resolution images are restored to the original im-ages.Comparisons between the original algorithms and our al-gorithm show that the proposed algorithm can effectively re-duce the matching time,screen out the mismatches,and im-prove the number of matches.