Abstract
We introduce a new complexity regularization method for image denoising and explore the use of sophisticated complexity penalties. We have found improvements of the order of 2 dB in reconstructed image mean-squared error over existing complexity-regularized estimators.
| Original language | English (US) |
|---|---|
| Pages | 370-373 |
| Number of pages | 4 |
| State | Published - Dec 1 1997 |
| Event | Proceedings of the 1997 International Conference on Image Processing. Part 2 (of 3) - Santa Barbara, CA, USA Duration: Oct 26 1997 → Oct 29 1997 |
Other
| Other | Proceedings of the 1997 International Conference on Image Processing. Part 2 (of 3) |
|---|---|
| City | Santa Barbara, CA, USA |
| Period | 10/26/97 → 10/29/97 |
ASJC Scopus subject areas
- Hardware and Architecture
- Computer Vision and Pattern Recognition
- Electrical and Electronic Engineering