Multicriteria Image Thresholding Based on Multiobjective Particle Swarm Optimization

This paper deals with using the MultiObjective Particle Swarm Optimization (MOPSO) [5] metaheuristic for optimally thresholding medical images. Two criteria are optimized: the interclass variance) and the Shannon entropy. The process generates a Pareto front with various segmentations, leaving the final choice to the user. The paper will, first, describe MOPSO algorithm and then, focus on obtained results.