RGB image multi-thresholding based on Kapur's entropy — A study with heuristic algorithms

In this paper, heuristic algorithm assisted multi-thresholding is attempted for standard RGB test images using Kapur's entropy. The motivation of this work is to present a detailed assessment among the Brownian search based heuristic approaches, such as bacteria foraging optimization, firefly algorithm, and cuckoo search. Proposed technique is implemented and validated on frequently used benchmark images and the performances of heuristic algorithms are validated by means of widely adopted quality measures accessible from the literature. From the experimental study, it is observed that, Brownian search based heuristic algorithms are efficient obtaining better image quality measures.