Fast Image Segmentation Based on Chaos Optimization and Recurring for 2-D Tsallis Entropy

The two-dimensional (2-D) maximum Tsallis entropy method takes advantage of the spatial neighbor information with using the 2-D histogram of the image and has a controllable parameter, so it often gets ideal segmentation results even when the image signal noise ratio (SNR) is low. However, its time-consuming computation is often an obstacle in real time application systems. In this paper, a fast image segmentation algorithm based on recurring and chaos optimization algorithm (COA) for 2-D Tsallis entropy is presented. Firstly, the traditional COA is improved, and then the improved COA, which can get global solution with lower computational load in the process of solving the 2-D maximum Tsallis entropy problem, is combined with recurring with the stored matrix variables to greatly reduce computational cost. Experimental results show the proposed approach can get better segmentation results with less computation cost.