Improved 2-D Tsallis Entropy Thresholding Based on Chaos Optimization

Owing to considering the distribution of the gray information and the spatial neighbor information with using the two-dimensional (2-D) histogram of the image, The 2-D maximum Tsallis entropy(2DMTE) method often gets better segmentation results, and owing to a controllable parameter, it has better flexibility than other 2-D entropy methods. However, its performance is sensitive to its parameter and it is time-consuming. Choosing the parameter and the much computation cost are often an obstacle in real time application systems. In this paper, improved image segmentation based on chaos optimization for 2-DTsallis entropy is presented. Firstly, the parameter of Tsallis entropy’s method is changed into two parameters, and the central value of the image cumulative distribution is obtained, then according to it, the parameters are selected adaptively, finally, the chaos optimization algorithm (COA) is applied to the process of solving the 2-D maximum Tsallis entropy problem to reduce much computation time. Experimental results show the proposed approach can get much better segmentation result than the previous 2-D thresholding methods with less computation cost.