Multi-Threshold Level Set Model for Image Segmentation

Abstract

A multi-threshold level set model for image segmentations is presented in the paper. The Multi-threshold level set formulation uses a speed function for stopping the locations of the active contours. The speed function with multiple thresholds is designed for detecting boundaries of multiple regions of interest (ROI) in images. These thresholds can be automatically obtained by Fuzzy C-means method. The experimental results show that the proposed method is able to capture boundaries of multiple regions of interest.