Abstract
We discover a new relationship between two seemingly different image modeling methodologies; the Besov space theory and the wavelet-domain statistical image models. Besov spaces characterize the set of real-world images through a deterministic characterization of the image smoothness, while statistical image models capture the probabilistic properties of images. By establishing a relationship between the Besov norm and the normalized likelihood function under an independent wavelet-domain generalized Gaussian model, we obtain a new interpretation of the Besov norm which provides a natural generalization of the theory for practical image processing. Based on this new interpretation of the Besov space, we propose a new image denoising algorithm based on projections onto the convex sets defined in the Besov space. After pointing out the limitations of Besov spaces, we propose possible generalizations using more accurate image models.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings of SPIE - The International Society for Optical Engineering |
| Publisher | Society of Photo-Optical Instrumentation Engineers |
| Pages | 489-501 |
| Number of pages | 13 |
| Volume | 3813 |
| State | Published - 1999 |
| Event | Proceedings of the 1999 Wavelet Applications in Signal and Image Processing VII - Denver, CO, USA Duration: Jul 19 1999 → Jul 23 1999 |
Other
| Other | Proceedings of the 1999 Wavelet Applications in Signal and Image Processing VII |
|---|---|
| City | Denver, CO, USA |
| Period | 7/19/99 → 7/23/99 |
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Condensed Matter Physics