On image denoising algorithm based on the grey system theory

A new method of the noise detection and a new adaptive weighted filter, based on the grey system theory and the mean image, are proposed in this paper. At first, the grey coefficients of incidence matrix are calculated between the noise image and the mean image and the noise spots are distinguished according to the relations between the grey coefficients of incidence matrix and the threshold value TH. Moreover, taking the pixels of the mean image which the noise spot corresponds as the center, the grey prediction model is built according to its near pixels on 3×3 the template. Then, the values of the noise spot can be replaced by the first forecasting value of the grey model. Finally, the simulation testing has been carried on under the different noise level, and the denoising effect has been evaluated by using the signal-to-noise ratio, Peak Signal to Noise Ratio and the mean error objectively. The result shows that the proposed method may reduce the image fuzziness, preserve the integrity of edge and detail information, and has the good denoising effect.