This article describes how multilevel thresholding image segmentation is a process used to partition an image into well separated regions. It has various applications such as object recognition, edge detection, and particle counting, etc. However, it is computationally expensive and time consuming. To alleviate these limitations, nature inspired metaheuristics are widely used to reduce the computational complexity of such problem. In this article, three cellular metaheuristics namely cellular genetic algorithm (CGA), cellular particle swarm optimization (CPSO) and cellular differential evolution (CDE) are adapted to solve the multilevel thresholding image segmentation problem. Experiments are conducted on different test images to assess the performance of the cellular algorithms in terms of efficiency, quality and stability based on the between-class variance and Kapur's entropy as objective functions. The experimental results have shown that the proposed cellular algorithms compete with and even outperform existing methods for multilevel thresholding image segmentation.
