Abstract
We present a new diffusion method for noise reduction and feature preservation.Presently, denoising methods commonly use a first-order derivative to detect edges inorder to achieve a good balance between noise removal and feature preserving.However, if edges are partly lost to a certain extent or contaminated severely bynoise, these methods may not be able to detect them and thus fail to preserve variousfeatures in images. To overcome this problem, we propose a new and moresophisticated feature detector by combining first- and second-order derivatives for anonlinear anisotropic diffusion model. Numerical experiments show that the newdiffusion filter outperforms many popular filters for denoising images containingedges, blobs and ridges and textures made of these features.
Original language | English |
---|---|
Title of host publication | BMVC 2011 - Proceedings of the British Machine Vision Conference |
Publisher | BMVA Press |
Number of pages | 11 |
DOIs | |
Publication status | Published - Sept 2011 |
Event | 22nd British Machine Vision Conference - Dundee, United Kingdom Duration: 29 Aug 2011 → 2 Sept 2011 http://www.bmva.org/bmvc/2011/bmvcproceedings.html |
Conference
Conference | 22nd British Machine Vision Conference |
---|---|
Abbreviated title | BMVC 2011 |
Country/Territory | United Kingdom |
City | Dundee |
Period | 29/08/11 → 2/09/11 |
Internet address |