多匹配基元集成的多视影像密集匹配方法.docx

第42卷第5期 第42卷第5期 测绘学报 V01．42，NO．5 2013年10月 Acta Geodaetica et Ca rtog raphica Si nica OCi．，2013 WANG Jingxue，ZHU Qing，WANG Weixi．A Dense Matching Algorithm of Multi View Image Based the Integrated Multiple Matching Primitives[-J]．Aeta Geodaetica Cartographica Sinica，2013，42(5)：691—698，(王竞雪，朱庆，王伟玺．多匹配基元集成的多视 影像密集匹配方法EJ]．测绘学报，2013，42(5)：691—698．) 多匹配基元集成的多视影像密集匹配方法 王竞雪1，朱 庆2，王伟玺3 1．辽宁工程技术大学测绘与地理科学学院，辽宁阜新123000；2．西南交通大学地球科学与环境工程学院，四川成都 610031；3．深圳市国土资源创新研究中心，广东深圳518034 A Dense Matching AIgo rithm of MuIti-view Image Based or；the Integ rated MuIiipie Matchina Primitives WANG Jingxuel，ZHU Qin92，WANG Weixi3 1．School of Geomatics，Liaonlng TechnicaI Unive rsity，Fuxin 123000，China；2．FacuIty of Geosciences and Envi rnentaI Engineering，Southwest Jiaotong University，Chengdu 610031，China；3．Shenzhen Innovation Research Cent of Land and Resou rces．Shenzhen 518034，Chlna Abst ract：Aiming at the occlusion p roblem of multi—view image matching，a new dense matching aIgo rithm was p resented by integ rating the featupoints in image object and plane elements in the space object·which is based the space plane divided in reguIa g rids．Fi rstly，it reaIized the simultaneously matching of featu points and plane elements by the constrains of the p rojection ranges of featu points in muIti—view image and the positions of plane elements．and provided reliable initiaI DSM fo the next dense matching．Then it ried the dense matching of regula distribution plane elements by combining the vertical line locus method and the height—based occlusion detection method，and densitied the initiaI matching results．FinaIly，the vaIidity of the aIgo rithm p roposed in this pape was verified by the experiments using fou UItraCamX(UCX)digitaI riaI images． Key wo rds：multi—view image matching；featu point in image object；plane element in space object；occlusion detection：ve rtIcaI Iine IOCUS 摘要：针对多视影像密集匹配中的遮挡问题，