Texture Features Analysis using Gray Level Co-occurrence Matrix for Abnormality Detection in Chest CT Images

Authors

  • Faleh H. Mahmood Remote sensing Research Unit, College of Science, University of Baghdad, Baghdad, Iraq
  • Wafaa A. Abbas College of Pharmacy, University of Baghdad, Baghdad, Iraq

Keywords:

Texture feature, Co-occurrence matrix, CT-scan, statistical feature

Abstract

Texture is an important characteristic for the analysis of many types of images because it provides a rich source of information about the image. Also it provides a key to understand basic mechanisms that underlie human visual perception. In this paper four statistical feature of texture (Contrast, Correlation, Homogeneity and Energy) was calculated from gray level Co-occurrence matrix (GLCM) of equal blocks (30×30) from both tumor tissue and normal tissue of three samples of CT-scan image of patients with lung cancer. It was found that the contrast feature is the best to differentiate between textures, while the correlation is not suitable for comparison, the energy and homogeneity features for tumor tissue always greater than its values for normal tissue.

Downloads

Download data is not yet available.

Downloads

Published

2023-02-08

Issue

Section

Remote Sensing

How to Cite

Texture Features Analysis using Gray Level Co-occurrence Matrix for Abnormality Detection in Chest CT Images. (2023). Iraqi Journal of Science, 57(1A), 279-288. https://ijs.uobaghdad.edu.iq/index.php/eijs/article/view/9324

Similar Articles

1-10 of 2658

You may also start an advanced similarity search for this article.