Section: New Software and Platforms

OWF

Keywords: Image filter - Image processing - Statistics

Functional Description: The OWF software enables to denoise images corrupted by additive white Gaussian noise. In the line of work of the Non-Local means and ND-SAFIR algorithms, this adaptive estimator is based on the weighted average of observations taken in a neighborhood with weights depending on the similarity of local patches. The idea is to compute adaptive weights that best minimize an upper bound of the pointwise L2 risk. The spatially varying smoothing parameter is automatically adjusted to the image context. The proposed algorithm is fast and easy to control and is competitive when compared to the more sophisticated NL-means filters.

• Participants: Qiyu Jin, Ion Grama, Quansheng Liu and Charles Kervrann

• Partner: University of Bretagne-Sud

• Contact: Charles Kervrann

• Publication: Non-local means and optimal weights for noise removal

• URL: http://serpico.rennes.inria.fr/doku.php?id=software:owf