Multilevel Thresholding for Image Segmentation Using Cricket Chirping Algorithm
S. Siva Sathya, Jonti Deuri · 2018
This chapter proposes a multi-level thresholding image segmentation method, based on the cricket chirping algorithm combined with Kapur&s;s entropy criterion method and Otsu&s;s between-class variance method. It discusses experimental results after testing the proposed method over a set of benchmark images and performing the comparison with its counterpart. A cricket is an insect similar to a grasshopper, with a flattened body that makes a sound that is known as chirping. Based on this chirping behavior of the cricket, a new cricket chirping algorithm (CCA) has been proposed by J. Deuri and S. S. Sathya. Kapur&s;s method is based on the entropy and the probability distribution of the image histogram; it is also known as the entropy criterion method. The between-class variance is a non-parametric and unsupervised technique for thresholding proposed by Otsu, one that employs the maximum variance value of the different classes as a criterion to segment the image.