High-dimensional multi-level image thresholding using self-organizing migrating algorithm

Abstract

Multi-level image thresholding is a common approach to image segmentation for which population-based metaheuristic algorithms present an interesting alternative to conventional methods that are based on a exhaustive search. In this paper, we propose a novel multi-level image thresholding algorithm based on the Self-Organizing Migrating Algorithm (SOMA), in particular SOMA Team To Team Adaptive (SOMA T3A), a recent variant of SOMA, and an entropy-based fitness function. We evaluate our algorithm on a set of benchmark images on high-dimensional search spaces and with regards to fitness function value and peak signal-to-noise ratio (PSNR). Experimental results demonstrate excellent thresholding performance and our algorithm to outperform nine other state-of-the-art metaheuristics.