A novel method for filtering of Gaussian colored noise in images with wavelet transform

Based on the statistical properties of the colored noise in wavelet domain and the whitening property of wavelet transform, we present a novel method to filter colored noise efficiently. The proposed method treats every detail subband in wavelet domain as a regular image with white noise, and filters the noise using the threshold value algorithm by iteratively performing wavelet decomposition. The image polluted by colored noise is then denoised by doing inverse transform. The method is independent of the correlation parameter of the colored noise. Our simulation results indicate that the proposed method is able to achieve close or better performance in filtering the colored noise with significantly reduced computation cost than existing approaches, and it is also applicable to reduce white noise.