Novel Image Denoising Method Based on Discrete Fractional Orthogonal Wavelet Transform
Wang You-re · Dianzi xuebao · 2014
The fractional wavelet transform extend the analysis method of wavelet transform about time-frequency domain to time-fractional frequency domain,can characterize signal features in time and fractional-frequency domain.Based on the theory of multiresolution analysis(MRA)of the fractional wavelet,the new forms about coefficient decomposition and reconstruction of discrete fractional wavelet transform(DFRWT)are given and made two-dimensional expansion.According to the feature of subband coefficient energy of image based on DFRWT varies with different p order,a novel image threshold denoising method based on DFRWT is proposed.The method in conditions of keeping the energy of low frequency subband for absolute great value,appropriate to raise the distribution of energy percentage,more beneficial for image noise suppression.The experimental results show that the performance of this method is better than the traditional wavelet threshold denoising method both in vision effect and peak signal to noise ratio.