February 29, 2008


Speaker: Jung-Ha An

Title: Gamma-Convergence Approximation to Region Based Piecewise Smooth Medical Image Segmentation

Abstract:  Despite many research efforts, accurate extraction of structures of interest still remains a difficult issue in many medical imaging applications. This is particularly the case in magnetic resonance (MR) images where image quality depends highly on the acquisition protocol.  A couple of variational region based algorithms which are able to deal with spatial perturbations of the image intensity directly are introduced in this talk.  The first model is obtained by minimizing an energy function which depends on a modified Mumford-Shah algorithm with numerical applications to simulated brain MR images.  In the second model, image segmentation is obtained by using a Gamma-Convergence approximation for a multi-scale piecewise smooth model.  The presented model is implemented efficiently using recursive Gaussian convolutions. Numerical experiments on 2-dimensional human liver MR images show that the model compares favorably to existing methods.