Image denoising method based on conversion coefficient statistical property

Abstract

The invention discloses an image denoising method based on conversion coefficient statistical property. High-frequency sub-bands, obtained after non-subsample contourlets are decomposed, of a noisy image are extracted, the statistical property of coefficients is depicted through Weibull distribution, modeling is performed through average cone rate, three relations between the non-subsample controurlet sub-bands are made full use of through a novel HMT, and the high-frequency sub-bands are denoised. Because the statistical property of the conversion coefficients is described more accurately through Weibull distribution and combined measurement of the average cone rate serves as a hidden state, the novel HMT is built through various relations between coefficient scales, in the scales and between directions. Thus, information and noise can be better recognized through the method, and the visual effect of the image is remarkably improved.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to an image denoising method based on the statistical characteristics of transform coefficients with reasonable denoising time and ideal denoising effect. Background technique [0002] Images are often polluted by various noises during transmission and acquisition, such as Gaussian white noise in optical images. The existence of noise will greatly reduce the resolution of the original image, which will seriously affect the subsequent advanced image processing, such as image registration, image segmentation, object classification, etc. Therefore, image denoising has always been the research focus in the field of computer vision and image processing, and has become a hot spot in international academic research. [0003] Image denoising is to perform a series of decomposed transformations on the image containing noise interference and then perform denoising processing, and then reconst...

AI Technical Summary

Problems solved by technology

But unfortunately, there are still many insurmountable problems in the existing denoising methods. For example, the bilateral filtering method cannot deal with Speckle noise, and often makes the image too smooth; the non-local mean method has two major defects: first, the objective quality and The visual effect is worse than other denoising methods; second, compared to other denoising algorithms, the computational complexity is , the computational complexity of the nonlocal algorithm is , where n is the size of the image; the conditional random field method also has two major drawbacks: first, the calculation of the energy function of the conditional random field must be feasible, but, in the real world, finding the global minimum for most energy functions is a NP problem; second, it is difficult to find an energy function with a global minimum in the desired solution; the anisotropic diffusion method is too smooth and the boundary is too sharp, so that a lot of texture information is lost

Images