文献类型：会议论文

英文题名：An Image Segmentation Method for Chinese Paintings by Combining Deformable Models with Graph Cuts

作者：He, Ning[1];Lu, Ke[2]

第一作者：何宁

通讯作者：He, N[1]

机构：[1]Beijing Union Univ, Sch Informat, Beijing 100101, Peoples R China;[2]Grad Univ Chinese Acad Sci, Coll Comput & Commun Engn, Beijing 100049, Peoples R China

第一机构：北京联合大学智慧城市学院

通讯机构：[1]corresponding author), Beijing Union Univ, Sch Informat, Beijing 100101, Peoples R China.|[1141734]北京联合大学智慧城市学院;[11417]北京联合大学;

会议论文集：International Conference on Ergonomics and Health Aspects of Work with Computers (EHAWC)/14th International Conference on Human-Computer Interaction (HCI)

会议日期：JUL 09-14, 2011

会议地点：Orlando, FL

语种：英文

外文关键词：Graph cuts; Deformable Model; Traditional Chinese Painting (TCP); Belief Propagation (BP); Po-Occurrence Matrix

摘要：In recent years researchers have developed many graph theory based algorithms for image setmentation. However, previous approaches usually require trimaps as input, or consume intolerably long time to get the final results, and most of them just consider the color information. In this paper we proposed a fast object extraction method. First it combines deformable models information with explicit edge information in a graph cuts optimization framework. we segment the input image roughly into two regions: foreground and background. After that, we estimate the opacity values for the pixels nearby the foreground/background border using belief propagation (BP). Third, we introduce the texture information by building TCP images' co-occurrence matrices. Experiments show that our method is efficient especially for TCP images.