Share Email Print
cover

Proceedings Paper

Learning sparse wavelet codes for natural images
Author(s): Bruno A. Olshausen; Phil Sallee; Michael S. Lewicki
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

We show how a wavelet basis may be adapted to best represent natural images in terms of sparse coefficients. The wavelet basis, which may be either complete or overcomplete, is specified by a small number of spatial functions which are repeated across space and combined in a recursive fashion so as to be self-similar across scale. These functions are adapted to minimize the estimated code length under a model that assumes images are composed as a linear superposition of sparse, independent components. When adapted to natural images, the wavelet bases become selective to different spatial orientations, and they achieve a superior degree of sparsity on natural images as compared with traditional wavelet bases.

Paper Details

Date Published: 4 December 2000
PDF: 8 pages
Proc. SPIE 4119, Wavelet Applications in Signal and Image Processing VIII, (4 December 2000); doi: 10.1117/12.408604
Show Author Affiliations
Bruno A. Olshausen, Univ. of California/Davis (United States)
Phil Sallee, Univ. of California/Davis (United States)
Michael S. Lewicki, Carnegie Mellon Univ. (United States)


Published in SPIE Proceedings Vol. 4119:
Wavelet Applications in Signal and Image Processing VIII
Akram Aldroubi; Andrew F. Laine; Michael A. Unser, Editor(s)

© SPIE. Terms of Use
Back to Top