Three-Level Gray-Scale Images Segmentation using Non-extensive Entropy

The segmentation of images into meaningful and homogenous regions is a crucial step in many image analysis applications. In this paper, we present a three-level thresholding method for image segmentation. The method is based on maximizing non-extensive entropy. After segmentation, the output image will consist of three homogenous regions, namely, dark, gray and white. The threshold value for each region is decided by maximizing an extended form of Tsallis entropy. To improve the performance of the proposed algorithm, an efficient and computationally fast method for initializing the search for maximum entropy is also presented. Results obtained using the proposed algorithm are compared with those obtained using Shannon entropy.