Title:Image Segmentation Using a Hybrid Genetic Algorithm with Tabu List for Maximum Tsallis Entropy Thresholding Authors: L. Jubair Ahmed ; A. Ebenezer Jeyakumar DOI: Aff: Department of Electronics & Communication Engineering,
Anna University, India, Author Email: email@example.com Keywords: Image Segmentation; Maximum Tsallis Entropy Thresholding; Comprehensive Learning PSO; Ant Colony Optimization URLs:ABSTRACT-HTML | FULLTEXT-PDF | Abstract:Background: Image thresholding is an important technique for image processing and pattern recognition. Many thresholding techniques have been proposed in the literature. Objective: In this paper to compute optimum thresholds for Maximum Tsallis entropy thresholding (MTET) model, a new hybrid algorithm is proposed based on the fusion of a Genetic Algorithm with the Tabu Search method. In this new theory a system dependent parameter ‘q’ measuring the degree of non-extensivity is introduced. The q parameter in the Tsallis entropy is used as an adjustable value, which plays an important role as a tuning parameter in the image segmentation. Thus replacing the traditional maximum entropy thresholding (MET) with a maximum Tsallis entropy. This new multilevel thresholding technique is called hybrid genetic algorithm (HGA) and tabu search algorithm for MTET. Results: Experimental results over multiple images with different range of complexities validate the efficiency of the proposed technique with regard to segmentation accuracy, speed, and robustness in comparison with other techniques reported in the literature. Conclusion: This article presents an extensive study on the application of a hybrid algorithm integrating two powerful metaheuristic techniques for multilevel thresholding for image segmentation problem. The proposed method was evaluated on various types of images, and the experimental results show the efficiency and the feasibility of the proposed method on the real images. It is demonstrated that the simulation results obtained using the hybrid HGA & tabu search method are superior to that of ABC, PSO and Genetic Algorithm methods in terms of producing quality thresholds. The validity and stability of the method is justified both qualitatively and quantitatively.