Multi-level Thresholding Segmentation Approach Based on Spider Monkey Optimization Algorithm

Image Segmentation is an open research area in which Multi-level thresholding is a topic of current research. To automatically detect the threshold, histogram-based methods are commonly used. In this paper, histogram-based bi-level and multi-level segmentation are proposed for gray scale image using spider monkey optimization (SMO). In order to maximize Kapur's and Otus's objective functions, SMO algorithm is used. To test the results of SMO algorithm, we use standard test images. The standard images are pre-tested and Benchmarked with Particle Swarm Optimization (PSO) Algorithm. Results confirm that new segmentation method is able to improve upon result obtained by PSO in terms of optimum threshold values and CPU time.