Optimal multilevel thresholding selection for brain MRI image segmentation based on adaptive wind driven optimization

Medical image processing is one of the real research regions in the most recent four decades. Numerous researchers have contributed very great algorithms and reported outstanding results. In this paper, adaptive wind driven optimization (AWDO) based multilevel thresholding is implemented for the segmentation Magnetic Resonance Image (MRI) brain images. The axial T2-weighted MRI brain images are considered for image segmentation. The effectiveness of the AWDO for multilevel thresholding of MR images is yet to be explored, and this paper presents humble contribution in this context. The optimal multilevel thresholding is found by maximizing the very popular objectives such as between class variance (Otsu method) and Kapur's entropy. The efficiency of proposed approach was compared with the outcomes of existing algorithms like RGA, GA, Nelder-Mead simplex, PSO, BF and ABF. To check the effectiveness of the proposed algorithm, experimental results are analyzed quantitatively and qualitatively. The results showed the superiority of the proposed approach regarding better segmentation results. (C) 2018 Elsevier Ltd. All rights reserved.