Feature quarrels: the dempster-shafer evidence theory for image segmentation using a variational framework

摘要: Image segmentation is the process of partitioning an image into at least two regions. Usually, active contours or level set based methods combine different feature channels, arising from color distribution, texture scale information, in energy minimization approach. In this paper, we integrate Dempster-Shafer evidence theory to fuse information (and resolve conflicts) channels. They are further combined with a smoothing term and applied signed distance function evolving contour. several experiments demonstrate properties advantages using segmentation.