{"title":"Experimental sensitivity analysis of image matching by relaxation labelling","authors":"N.M.F. de Oliveira, W. Waldmann","doi":"10.1109/CYBVIS.1996.629466","DOIUrl":null,"url":null,"abstract":"Presents an experimental evaluation of the sensitivity of a relaxation labelling scheme for image matching to variations in tuning parameters usually set by an operator. Information concerning camera displacement is employed to reduce the complexity and ambiguities that arise when matching regions with significant contrast. The use of displacement information is seen to yield results more robust than those observed without any use of displacement information whatsoever. The experiment has been conducted with real images and the observed robustness encourage further use of the algorithm in vision systems that require the fusion of visual and inertial data, such as the one now being developed in the Active Computer Vision and Perception Laboratory ITA/INPE.","PeriodicalId":103287,"journal":{"name":"Proceedings II Workshop on Cybernetic Vision","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1996-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings II Workshop on Cybernetic Vision","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CYBVIS.1996.629466","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Presents an experimental evaluation of the sensitivity of a relaxation labelling scheme for image matching to variations in tuning parameters usually set by an operator. Information concerning camera displacement is employed to reduce the complexity and ambiguities that arise when matching regions with significant contrast. The use of displacement information is seen to yield results more robust than those observed without any use of displacement information whatsoever. The experiment has been conducted with real images and the observed robustness encourage further use of the algorithm in vision systems that require the fusion of visual and inertial data, such as the one now being developed in the Active Computer Vision and Perception Laboratory ITA/INPE.