Extending contour level set model for multi-class image segmentation with Application to Breast Thermography Images

•ECLS model is proposed to segment a target region in the multi-region images.•It can extract target regions in images with different levels of complicated intensity. It is not sensitive to initial contour location.•Simulation results indicate the remarkable advantages of the proposed method in terms of accuracy and computational complexity than other level set methods.•In the proposed algorithm, determination of the number of image segments is not crucial; therefore, it can extract regions with high temperature despite their complicity in the hot regions and the background.•It has more accurate result in hot region extraction, than other comparative methods.