Paper Details: Downloads: 474
Serial Number: P1150836349
Title: Texture Characterization of CT Images Based on Ridgelet Transform
Authors: V.R. Ratnaparkhe and R.R.Manthalkar and Y.V.Joshi
Abstract: As the human vision system cannot distinguish the broad range of gray values that a computer visual system can, computerized image analysis has been used to obtain quantitative information from images. This article presents an approach of extraction of texture features of medical images. Images acquired from CT scan imaging modality have been used for the work. Classification of tissues in medical images based on texture features has been reported here. It involves two phases. During first phase, the texture features of selected ROIs are determined and are classified during the next phase. Ridgelet based techniques have been implemented for texture features. Typically Percentage Correct Classification parameter is expressed for comparative study of Ridgelet based energy and entropy features in these techniques. The analysis based on confusion matrix and correct classification for two-class and multi-class classification has been presented.
Keywords: Texture, Ridgelet Transform, Classification
Journal/Conference: International Journal of Graphics, Vision and Image Processing
Volume: 8
Issue: 5
Submission Date: 9/3/2008 12:00:00 AM
Review Date: 10/5/2008 12:00:00 AM
Publishing Date: 11/4/2008 12:00:00 AM
Article Downloads: 474
Download:

Facebook