Fuzzy 2-partition entropy threshold selection based on Big Bang–Big Crunch Optimization algorithm

• Main Authors: Baljit Singh Khehra, Amar Partap Singh Pharwaha, Manisha Kaushal
• Format: Article
• Language: English
• Published: Elsevier 2015-03-01
• Series: Egyptian Informatics Journal
• Subjects: Big Bang–Big Crunch Optimization; Biogeography-based Optimization; Fuzzy 2-partition entropy; Optimal threshold; Image segmenting
• Online Access: http://www.sciencedirect.com/science/article/pii/S1110866515000080

The fuzzy 2-partition entropy approach has been widely used to select threshold value for image segmenting. This approach used two parameterized fuzzy membership functions to form a fuzzy 2-partition of the image. The optimal threshold is selected by searching an optimal combination of parameters of the membership functions such that the entropy of fuzzy 2-partition is maximized. In this paper, a new fuzzy 2-partition entropy thresholding approach based on the technology of the Big Bang–Big Crunch Optimization (BBBCO) is proposed. The new proposed thresholding approach is called the BBBCO-based fuzzy 2-partition entropy thresholding algorithm. BBBCO is used to search an optimal combination of parameters of the membership functions for maximizing the entropy of fuzzy 2-partition. BBBCO is inspired by the theory of the evolution of the universe; namely the Big Bang and Big Crunch Theory. The proposed algorithm is tested on a number of standard test images. For comparison, three different algorithms included Genetic Algorithm (GA)-based, Biogeography-based Optimization (BBO)-based and recursive approaches are also implemented. From experimental results, it is observed that the performance of the proposed algorithm is more effective than GA-based, BBO-based and recursion-based approaches.