Bacterial Foraging Optimization Algorithm with Varying Population for Entropy Maximization based Image Segmentation

In this paper, a new variant of bacterial foraging optimization (BFO) algorithm, called bacterial foraging with varying population (named CBFVPA) is proposed for bi-level thresholding based gray scale image segmentation. The work shows how CBFVPA can be effectively utilized for fuzzy entropy maximization and how it can improve upon the performance of classical BFO (named as CBFOA) utilized for solving similar problems. In contrast to CBFOA, where a fixed population of bacteria is utilized, the basic essence of CBFVPA is that the population size undergoes variation through the phases of chemotaxis, metabolism, elimination and quorum sensing, in each iteration. The proposed algorithm has been employed on several benchmark gray scale images and the segmentation performances are computed in terms of a popular performance index, called uniformity factor. The performances show that CBFVPA is able to provide an overall, superior performance compared to that of CBFOA.