A Novel Image Segmentation Based on Clustering and Population-Based Optimisation

Mousavirad, Seyed Jalaleddin; Schaefer, Gerald; Ebrahimpour-Komleh, Hossein; Korovin, Iakov

Image segmentation is an essential step in image processing and computer vision with many image segmentation algorithms having been proposed in the literature. Among these, clustering is one of the prominent approaches to achieve segmentation. Traditional clustering algorithms have been used extensively for this purpose, although they have disadvantages such as dependence on initialisation conditions and a tendency to find only local optima. To overcome these disadvantages, population-based metaheuristic algorithms can be applied. In this paper, we propose a novel clustering algorithm based on human mental search (HMS) for image segmentation. HMS is a relatively new population-based metaheuristic inspired from the manner of searching in online auctions. HMS comprises three operators: mental search, which explores the neighbourhood of candidate solutions using Levy flight; grouping, which clusters candidate solutions; and moving candidate solutions towards a promising area. To verify the efficacy of the proposed algorithm, we conduct several experiments based on different criteria including mean cost function value, statistical analysis and image segmentation criteria. The obtained results confirm superior performance of our proposed algorithm compared to competitors.