Image Segmentation Combining Level Sets and Principal Component Analysis

In this thesis, we present an enhancement to the Stochastic Active Contour Sche~ne (STACS) [4] for Image Segmentation using Principal Component Analysis(PCA). STACS is a method developed for Segmentation of Cardiac Magnetic Resonance Imaging (MRI) images and is based on the level set method in which the contaur is driven by the minimization of a function of four terms-region based, edge based, shape prior, and curvature. STACS derives each of these forces from the original image that is to bc segmented. In our method, we perform PCA on the entire set of eight images of the same slice of the heart taken at different instants of time in the c~trdiac cycle and then segment each image separately. The various terms in the ene~yy functional in this new scheme are obtained from different principal components (Eigenvectors). Thus. modifying STA CS as explained, we improve it emphasizing each term in the energy functional with the help of the principal component that gives the most accurate result. We present experimental results with the proposed scheme and discuss how to extend the approach in future work.