期刊名称:International Journal of Computer Science and Network Security
印刷版ISSN:1738-7906
出版年度:2009
卷号:9
期号:8
页码:123-131
出版社:International Journal of Computer Science and Network Security
摘要:Active contours or snakes represent widely used methods in edge segmentation. Different, closed snake approaches have been proposed in the past but commonly lack of adequate capture ability and quality, noise robustness and convergence speed. In this paper a new closed approach namely ��Alternating Vector Field Convolution (AVFC)�� as well as a new force called ��Yukawa Alternating Vector Field Convolution (YAVFC)�� are proposed which minimize specific snake problems and enhance convergence speed. The comparison of both Gradient Vector Flow and Vector Field Convolution snakes with AVFC and YAVFC snakes determines the advantages of the new closed approaches.
关键词:Active contour; Snake; External force; Yukawa; Alternating vector field convolution