Gray Level Co-Occurrence Matrix Computation Based on Haar Wavelet
In this paper, a new computation for gray level co-occurrence matrix (GLCM) is proposed. The aim is to reduce the computation burden of the original GLCM computation. The proposed computation will be based on Haar wavelet transform. Haar wavelet transform is chosen because the resulting wavelet band...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
2007
|
| Online Access: | http://eprints.utm.my/id/eprint/8774/ http://doi.ieeecomputersociety.org/10.1109/CGIV.2007.45 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In this paper, a new computation for gray level co-occurrence matrix (GLCM) is proposed. The aim is to reduce the computation burden of the original GLCM computation. The proposed computation will be based on Haar wavelet transform. Haar wavelet transform is chosen because the resulting wavelet bands are strongly correlated with the orientation elements in the GLCM computation. The second reason is because the total pixel entries for Haar wavelet transform is always minimum. Thus, the GLCM computation burden can be reduced. The proposed computation is tested with the classification performance of the Brodatz texture images. Although the aim is to achieve at least similar performance with the original GLCM computation, the proposed computation gives a slightly better performance compare to the original GLCM computation.
|
|---|