Skip to Main Content
Skip Nav Destination
ASME Press Select Proceedings
International Conference on Computer Technology and Development, 3rd (ICCTD 2011)
By
Jianhong Zhou
Jianhong Zhou
Search for other works by this author on:
ISBN:
9780791859919
No. of Pages:
2000
Publisher:
ASME Press
Publication date:
2011

Texture retrieval is a significant branch of content based image retrieval (CBIR). In this paper, the nonsubsampled contourlet transform has been used for texture image retrieval. This transform is built upon nonsubsampled pyramid and nonsubsampled directional filter banks and provides a shift-invariant, directional, multi-resolution image representation. The statistical characteristics of these directional subbands' coefficients have been used as the feature vector for texture image classification. The performance of the proposed method is compared to other texture image retrieval methods such as Gabor filter, Contourlet and Complex Directional Filter Bank (CDFB) and the experimental results show that our proposed method improves retrieval rate and outperforms other approaches in this concept.

Key Words
1. Introduction
2. Nonsubsampled Contourlet Transform
3. Classification Method
4. Experimental Results
5. Summaries
References
This content is only available via PDF.
You do not currently have access to this chapter.
Close Modal

or Create an Account

Close Modal
Close Modal