Abstract: A novel denoising model for texture images was proposed,which combines the new multiscale geometric analysis tool-wave atoms and nonlocal total variation regularization scheme. This model well considers the advantages:wave atoms which has the ability of sparse representation of the oscillatory texture images and nonlocal TV version has the ability to handle better textures. Numerical experiments show that the proposed model significantly improves the SNR, performance of preserving details is much better than both only wave atoms threshold and nonlocal TV, therefore together with better visual effects.

Key words: Image denoising, Wave atoms, Textures, Nolocal TV, Threshold