IMAGE SEGMENTATION BY A DEFORMABLE CONTOUR MODEL INCORPORATING REGIONANALYSIS

Deformable contour models are useful tools for image segmentation. How ever, many models depend mainly on local edge-based image features to guide the convergence of the contour. This makes the models sensitive to noise and the initial estimate. Our model incorporates region-based image features to improve its convergence and to reduce its dependenc e on initial estimation. Computational efficiency is achieved by an op timization strategy, modified from the greedy algorithm of Williams an d Shah. The model allows a simultaneous optimization of multiple conto urs, making it useful for a large variety of segmentation problems.