Texture Characterization Using a Curvelet Based Descriptor

Authors:
Francisco Gómez;Eduardo Romero
Affiliations:
Bioingenium Research Group - Faculty of Medicine, University of Colombia, Bogotá, Colombia;Bioingenium Research Group - Faculty of Medicine, University of Colombia, Bogotá, Colombia
Venue:
CIARP '09 Proceedings of the 14th Iberoamerican Conference on Pattern Recognition: Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
Year:
2009

Citing 6
Cited 0

Filtering for Texture Classification: A Comparative Study

IEEE Transactions on Pattern Analysis and Machine Intelligence
New multiscale transforms, minimum total variation synthesis: applications to edge-preserving image reconstruction

Signal Processing - Image and Video Coding beyond Standards
A comparison of wavelet, ridgelet, and curvelet-based texture classification algorithms in computed tomography

Computers in Biology and Medicine
Texture Classification Using Three Circular Filters

ICVGIP '08 Proceedings of the 2008 Sixth Indian Conference on Computer Vision, Graphics & Image Processing
Statistical texture characterization from discrete wavelet representations

IEEE Transactions on Image Processing
Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Feature extraction from images is a key issue in image classification, image representation and content based image retrieval. This paper introduces a new image descriptor, based on the curvelet transform. The proposed descriptor captures edge information from the statistical pattern of the curvelet coefficients in natural images. The image is mapped to the curvelet space and each subband is used for establishing the parameters of a statistical model which captures the subband marginal distributions as well as the dependencies across scales and orientations of the curvelet. Finally, the Kullback***Leibler distance between the statistical parameters is used to measure the distance between images. We demonstrate the effectiveness of the proposed descriptor by classifying a set of texture images, and with a simple nearest neighbour classifier we obtained an accuracy rate of 87%.