Multilevel Image Thresholding Using Tsallis Entropy and Cooperative Pigeon-inspired Optimization Bionic Algorithm

Multilevel thresholding is a simple and effective method in numerous image segmentation applications. In this paper, we propose a new multilevel thresholding method that uses cooperative pigeon-inspired optimization algorithm with dynamic distance threshold (CPIOD) for boosting applicability and the practicality of the optimum thresholding techniques. Firstly, we employ the cooperative behavior in the map and compass operator of the pigeon-inspired optimization algorithm to overcome the “curse of dimensionality” and help the algorithm converge fast. Then, a distance threshold is added to maintain the diversity of the pigeon population and increase the vitality to avoid local optimization. Tsallis entropy is used as the objective function to evaluate the optimum thresholds for the considered gray scale images. Four benchmark images are applied to test the property and the stability of the proposed CPIOD algorithm and three other optimization algorithms in multilevel thresholding problems. Segmentation results of four optimization algorithms show that CPIOD algorithm can not only get higher quality segmentation results, but also has better stability.