{"title":"Shape decomposition based on erosion model","authors":"F. Kanehara, S. Satoh, T. Hamada","doi":"10.1109/PBMCV.1995.514677","DOIUrl":null,"url":null,"abstract":"W e present a shape decomposition method based o n e ros ion model t o extract convex parts f r o m a planar closed shape. Our approach i s based o n the curvature analysis of shape contours assuming that divisions of a shape generally occurs at concavities o n the contour, as if the sea gained o n the land along the coast. W e call this model as e ros ion model. The outline of our algorithm i s as follows. As a beginning, the vectorized shape contour i s extracted f r o m a silhouette image. Next , a n erosional vector which represents a force of erosion at each concavity i s computed, and then, a boundary which splits the shape i s decided. These processes are recursively performed until no division is available. Finally, some parts are merged if necessa y. T h e advantage of our approach i s that the meaningful convex part can be obtained ef ic ient ly and the decomposit ion results are theoretically invariant to translat ion, rotation and scaling.","PeriodicalId":343932,"journal":{"name":"Proceedings of the Workshop on Physics-Based Modeling in Computer Vision","volume":"21 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the Workshop on Physics-Based Modeling in Computer Vision","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PBMCV.1995.514677","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
W e present a shape decomposition method based o n e ros ion model t o extract convex parts f r o m a planar closed shape. Our approach i s based o n the curvature analysis of shape contours assuming that divisions of a shape generally occurs at concavities o n the contour, as if the sea gained o n the land along the coast. W e call this model as e ros ion model. The outline of our algorithm i s as follows. As a beginning, the vectorized shape contour i s extracted f r o m a silhouette image. Next , a n erosional vector which represents a force of erosion at each concavity i s computed, and then, a boundary which splits the shape i s decided. These processes are recursively performed until no division is available. Finally, some parts are merged if necessa y. T h e advantage of our approach i s that the meaningful convex part can be obtained ef ic ient ly and the decomposit ion results are theoretically invariant to translat ion, rotation and scaling.