A New Multi-phase Level Set Framework for 3D Medical Image Segmentation Based on TPBG
Gang Zheng, Yuanlu Li, Huinan Wang · 2005
In this paper, we propose a new multi-phase level set framework for 3D medical image segmentation to deal with the limitation of 2-phase segmentation algorithms using one level set. By developing the technique of painting the background with average gray level of the object (TPBG) and reusing the active contours model without edges [1] which is 2-phase one, we are able to obtain more than two segments (n-1 times for n phases, n>1). Following the philosophy of dichotomy, we illustrate the efficiency of the proposed framework by segmenting a 3D medical image into more than two segments step by step.