Image denoising based on wavelet transform and wiener filtering

How to remove the noise and meanwhile protect the details is a difficulty in image denoising. To rind a solution to the problem, a locally adaptive denoising algorithm in the wavelet domain is presented. Firstly, this algorithm applies wavelet transform to the corrupted image. Secondly, in high frequency part, the method deals the coefficients with wiener filtering on different subands and orientations. And in the low frequency domain, it leaves the coefficients without change. Finally, the inverse wavelet transform is performed on the modified coefficients. And then the method results in the denoised image. Experimental results showed that the method could achieve high quality image denoising. Among the traditional methods using the same type of wavelet filter, the proposed method produced the highest peak signal-to-noise ratio (PSNR). This algorithm is effective both in removing noise and in preserving the details.