Image compression using learned dictionaries by RLS-DLA and compared with K-SVD

The recently presented recursive least squares dictionary learning algorithm (RLS-DLA) is tested in a general image compression application. Dictionaries are learned in the pixel domain and in the 9/7 wavelet domain, and then tested in a straightforward compression scheme. Results are compared with...

Full description

Saved in:
Bibliographic Details
Main Authors: Skretting, Karl, Engan, Kjersti
Format: Conference Proceeding
Language:English
Subjects:
Dictionaries
dictionary learning
Discrete cosine transforms
Image coding
image compression
overcomplete dictionary
PSNR
RLS-DLA
sparse approximation
Training
Transform coding
Wavelet domain
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The recently presented recursive least squares dictionary learning algorithm (RLS-DLA) is tested in a general image compression application. Dictionaries are learned in the pixel domain and in the 9/7 wavelet domain, and then tested in a straightforward compression scheme. Results are compared with state-of-the-art compression methods. The proposed compression scheme using RLS DLA learned dictionaries in the 9/7 wavelet domain per forms better than using dictionaries learned by other methods. The compression rate is just below the JPEG 2000 rate which is promising considering the simple entropy coding used.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2011.5946782