Noise reduction of images with multiple subband transforms

T. Tanaka, L. Duval
2004 International Conference on Image Processing, 2004. ICIP '04.  
It is reported that the use of multiple number of subband transforms for thresholding-based denoising gains performance in the sense of the mean square error. In traditional thresholding-based methods, a noisy image is decomposed by linear transformation such as wavelets, FFT, and so on, and the transformed coe cients are hard-¡ soft-thresholded. In particular, it is well-known that wavelets work well for denoising. From the viewpoint that wavelets are in a class of subband transforms, we
more » ... ransforms, we propose a strategy in which multiple number of subband transforms are switched region by region, i.e. block by block. For reconstruction, the projection-based iterative method is used. Experimental results are pretty good and promising.
doi:10.1109/icip.2004.1419522 dblp:conf/icip/TanakaD04 fatcat:6jv2hgp6mrhj5dglcykkpz7cke