Vector-valued image segmentation using region based active contour
Zahra Shahvaran, Kamran Kazemi, Mohammad Sadegh Helfroush, Narjes Golshaeyan · 2013
In this paper we present our framework for object segmentation in vector-valued images with intensity non-uniformity and noise using region based level set method. The proposed framework extend the local Gaussian distribution fitting in which the local intensity is modeled by Gaussian distribution. Our experiments on synthetic images and simulated cerebral MR images show the advantages of the proposed method in comparison with applying scalar model.