Softcomputing techniques a promising boon to certain professional investigations on the classification and segmentation of computed tomography brain images

Abstract

Subsequent to the process of classification, the more prevalently used part in most of the applications of image processing and computer vision is the image segmentation. The entire study concerning the Computed Tomography (CT) holds image segmentation as a very essential or even an inevitable part in identifying the different kinds of tumor in the different levels. Once the classification of the parts or portions in the images as tumorous and non-tumorous is over, what follows next is the process of segmentation of the tumor regions in the CT images and it is the proposed methodology that takes the entire care of these both, classification and segmentation as well. For the purpose of classifying, the Support Vector Machine (SVM) with different kernels and optimization techniques is put into use. When it amounts to classification and optimization, the SVM classifier with Sequential Minimal Optimization (SMO) enjoys a clear predominance over the other methodologies in the analysis of classification process. Following the classification process, the Modified Region Growing (MRG) with threshold optimization fulfills the segmentation process. newline