Hybrid Optimization Algorithm for Multi-level Image Thresholding Using Salp Swarm Optimization Algorithm and Ant Colony Optimization

Abstract

The process of identifying optimal threshold for multi-level thresholding in image segmentation is a challenging process. An efficient optimization algorithm is required to find the optimal threshold and various nature inspired; evolutionary optimization algorithms are presented by the research community. However, to improve the performance in finding optimal threshold value and minimize the error, reduces the searching time a hybrid optimization algorithm is presented in this research work using salp swarm optimization and ant colony optimization algorithm. The ant colony optimization algorithm is used to enhance the exploration and exploitation characteristics of salp swarm optimization in finding optimal threshold for the given image. Experimentation using standard images validates the proposed model performance in comparison with traditional optimization algorithms like moth flame optimization, whale optimization algorithm, grey wolf optimization, artificial bee colony and bee foraging optimization algorithms. Proposed hybrid optimization outperformed in all parameters compared to traditional optimization algorithms and provides better optimal thresholds for the given input image.