Enhanced active shape model using evolutionary computation and its extension to RGB color space

This paper deals with an application of evolutionary computation to active shape model (ASM) for locating and analyzing objects an image. The ASM algorithm has proved to be a successful method for segmentation of objects in gray-scale images. However, a number of modifications and extensions to the basic ASM algorithm have been proposed to cope up with various limitations. The proposed method in this paper relies on two such extensions: (i) evolutionary algorithm for better local structure model; and (ii) incorporation of color information for solving the outlier problem in model fitting.