Multi-level thresholding is one of the essential approaches for image segmentation. Determining the optimal thresholds for multi-level thresholding needs exhaustive searching which is time-consuming. To improve the searching efficiency, a novel population based bee foraging algorithm (BFA) for multi-level thresholding is presented in this paper. The proposed algorithm provides different flying trajectories for different types of bees and takes both single-dimensional and multi-dimensional search aiming to maintain a proper balance between exploitation and exploration. The bee swarm is divided into a number of sub-swarms to enhance the diversity. A neighbourhood shrinking strategy is applied to mitigate stagnation and accelerate convergence. Experiments have been performed on eight benchmark images using between-class variance as the thresholding criterion. The performance of the proposed algorithm is compared with some state-of-art meta-heuristic algorithms. The results show that BFA is efficient and robust, produces excellent results with few control parameters, and outperforms other algorithms investigated in this consideration on most of the tested images. INDEX TERMS Bee foraging algorithm, image segmentation, meta-heuristic, multi-level thresholding.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.