Title: SVM classification of brain images from MRI scans using morphological transformation and GLCM texture features

Authors: R. Usha; K. Perumal

Addresses: Department of Computer Applications, School of Information Technology, Madurai Kamaraj University, Madurai, Tamilnadu, India ' Department of Computer Applications, School of Information Technology, Madurai Kamaraj University, Madurai, Tamilnadu, India

Abstract: This paper introduces a novel HTT-based GLCM texture feature extraction procedure for an automatic magnetic resonance images (MRI) brain image classification. The method has three phases: 1) hierarchical transformation technique (HTT); 2) texture feature extraction; 3) classification. The new proposed HTT method incorporates optimum disk-shaped mask selection, top-hat and bottom-hat morphological operations, and some mathematical operation for both image pre-processing and enhancement. The gray level co-occurrence matrix is computed to extract statistical texture features such as contrast, correlation, energy, entropy, and homogeneity from an image. And these extracted images features of co-occurrence matrix can very well be fed into support vector machine (SVM) for further MRI brain normal and abnormal image classification. The alternate approach of the HTT-based GLCM also compared with conventional GLCM texture feature extraction method.

Keywords: magnetic resonance images; MRI; classification; texture feature extraction; grey level co-occurrence matrix; support vector machine; SVM; top hat transform; bottom hat transform.

DOI: 10.1504/IJCSYSE.2019.098415

International Journal of Computational Systems Engineering, 2019 Vol.5 No.1, pp.18 - 23

Received: 29 May 2017
Accepted: 05 Aug 2017
Published online: 22 Mar 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article