Brain Mri Image Segmentation Using Simple Linear Iterative Clustering (SLIC) Segmentation With Superpixel Fusion

Abstract

Brain MRI image segmentation is a challenging task owing to the complex texture of the various parts of the brain.The goal of the segmentation is to divide the image into parts so that the classification algorithm can recognize the tumour present in the image.This paper presents a modified Simple Linear Iterative Clustering (SLIC) segmentation algorithm to segment the brain MRI image.The SLIC segmentation divides the brain MRI image into many superpixels.This algorithm uses both spatial intensity value and the pixel positions to form the superpixels.These superpixels are then fused with the neighbouring superpixels to form bigger regions in order to merge similar regions.The fusion operation is implemented to merge the superpixels which constitute the tumour in the image.This would help the classification algorithms to perform better.The proposed segmentation method is compared with existing techniques to analyse the performance.The existing algorithms are compared in terms of execution time, average deviation from and entropy.The experimental results prove that the proposed method segmented the brain MRI image better than the existing algorithms.