Image Denoising Based on Dilated Singularity Prior

In order to preserve singularities in denoising, we propose a new schemeby adding dilated singularity prior to noisy images. The singularities are detected bycanny operator firstly and then dilated using mathematical morphology for findingpixels “near” singularities instead of “on” singularities. The denoising results for pixelsnear singularities are obtained by nonlocal means in spatial domain to preserve singularitieswhile the denoising results for pixels in smooth regions are obtained by EMalgorithm constrained by a mask formed by downsampled spatial image with dilatedsingularity prior to suiting the sizes of the subbands of wavelets. The final denoised resultsare got by combining the above two results. Experimental results show that the schemecan preserve singularity well with relatively high PSNR and good visual quality.