Multi-level Image Thresholding based on Local Variance and Particle Swarm Optimization

Multi-level thresholding is a basic pre-processing technique in computer vision and pattern recognition tasks. Optimal threshold values classify pixel values into multiplecategories. The proposed method performs multi-level grayscale image thresholding based on local variance of pixelsand Particle Swarm Optimization (PSO). Local variancecontains foreground and background variances. The main idea of the proposed technique is combination between Sobelability, foreground and background variances in order to provide higher visual perception. This novelty causes betterexploration of search space in order to find appropriatethreshold values for region uniformity. Experimental results show the superiority of this approach in comparison with other thresholding approaches.