Multilevel image thresholding using relative entropy and Virus Optimization Algorithm

Abstract

One of the most popular techniques for image segmentation is known as multilevel thresholding. The main difference between multilevel and binary thresholding, is that the binary thresholding outputs a two-color image, usually black and white, while the multilevel thresholding outputs a grey scale image in which more details from the original picture can be kept. Two major problems with using the multilevel thresholding technique are: it is a time consuming approach, i.e., finding appropriate threshold values could take exceptionally long computational time; defining a proper number of thresholds or levels that will keep most of the relevant details from the original image is a difficult task. In this study a new approach based on the Kullback-Leibler information distance, also known as Relative Entropy, is proposed. The approach minimizes a mathematical model, which will determine the number of thresholds automatically. The optimization of the mathematical model is achieved by using a newly developed meta-heuristic named Virus Optimization Algorithm (VOA), where its performance is compared with Genetic Algorithm (GA). From the experiments performed in this study, the proposed method does not only provide good segmentation results but also its computational effort makes it a very efficient approach.