Active region segmentation of color images using neural networks

Abstract

Region segmentation of images is a well-known “ill-posed problem,” and a specific algorithm-like regularization seems to be available. In this paper, an active-region segmentation algorithm based on a regularization approach using the Hopfield neural network is proposed. Pyramid images are used to avoid local minima and to achieve fast convergence. Experimental results show that it is possible to segment images by using the minimization principle of the energy function of the Hopfield neural network. The active region segmentation algorithm is applied to a sequence of color images to track an object region that changes in appearance through complex and nonstationary background/foreground situations. © 1998 Scripta Technica, Syst Comp Jpn, 29(4): 1–10, 1998