【摘要】 提出了一种有利于图像压缩的小波图像去噪方法及一种小波系数校验方法。首先利用多个阈值在小波域进行硬阈值迭代去噪,然后利用小波子带的方向性对小波系数进行校验,并将保留的非零小波系数乘以一个比例系数,最后利用小波逆变换获得去噪图像。实验结果表明该方法的去噪效果优于常用方法,接近甚至超过BayesShrink方法。该方法可在有效去除图像噪声与保留较少的非零小波系数之间得到较好的折衷。更多还原

【Abstract】 An image denoising algorithm suitable for image compression and a wavelet coefficient verification method are proposed.At first,several thresholds are utilized to denoise image iteratively in the wavelet domain.Then retained nonzero wavelet coefficients are verified based on orientations of different wavelet subbands.And each nonzero wavelet coefficient is multiplied by a ratio coefficient.Finally inverse wavelet transform is performed to obtain the denoised image.Experimental results show that this algorithm outperforms conventional approaches.The denoising results of this algorithm are similar to and even superior to those of BayesShrink method.This approach can provide a good tradeoff between reducing image noise effectively and retaining a relatively small number of nonzero wavelet coefficients.更多还原