MINIMUM CROSS ENTROPY BASED IMAGE SEGMENTATION USING NEW OPTIMIZATION ALGORITHM

Abstract

Image segmentation is an essential advance for some picture investigation and preprocessing assignments. In
segmentation, minimum cross entropy (MCE) based multilevel thresholding is viewed as a viable improvement over the bilevel technique. Be that as it may, it is extremely tedious for continuous applications. In this paper, a quick limit
determination technique in light of bacterial foraging optimization (BFO) algorithm is proposed to accelerate the first MCE
edge strategy in picture division. BFO calculation is a recently evolved memetic meta-heuristic transformative algorithm with
great worldwide inquiry capacity. Exploratory outcomes contrasted and particle swarm optimization (PSO) and genetic
algorithm (GA) show that the BFO based thresholding can precisely acquire the worldwide ideal edge values with huge
abatement in the computational time and give better peak to signal noise ratio (PSNR) value and stability.