基于二进小波的图像去噪技术

Image denoising technique based on dyadic wavelet

图像在二进小波变换空间的表示是冗余的,同小波级数相比,基于二进小波变换的图像重建对于图像单个小波变换系数的依赖性下降。因此,在相同的误判概率下,基于二进小波变换的图像去噪效果好于基于小波级数变换的图像去噪效果,基于这个思想,将基于小波级数的图像去噪方法推广到基于二进小波变换的图像去噪,实验表明二进小波去噪可以有效地提高信噪比。

Image' s representation in dyadic wavelet domain is very redundant. Compared with wavelet seies reconstruction,image' s dyadic wavelet reconstruction dependency on the individual coefficients in transform domain will be decreased. Therefore,under the same misjudgment probability,image' s denosing efficiency of dyadic wavelet is better than that of wavelet series. Based on this idea, it is extended the image' s wavelet series denosing approaches to the image' dyadic wavelet denoising. Numerical experiments show that the dyadic wavelet denoising can significantly improve the signal-to-nosise rate.