基于小波包分析二次阈值去噪图像复原方法

提出了基于小波包分析的二次去噪复原的图像复原新方法。利用小波包分析在信号去噪中对图像信号高频部分更加灵活、更加精确的局部分析能力,针对低信噪比模糊图像,通过选取较小的阈值,交替地进行了两次预去噪处理和复原。采用PML算法,实验比较了小波包阈值一次去噪法和二次去噪法。实验表明,二次去噪法由于阈值取值较小,对退化图像信息的能量分布改变较小,可以获得比一次去噪法更加良好的复原效果。

Image restoration method by twice denoising based on wavelet packet analyzing

A new method for image restoration based on wavelet packet analysis is proposed by twice denoising.Using flexibility and the precise ability of local analysis in wavelet packet analysis during signal denoising to high frequency part of image signal,in view of low signal-to-noise choosing lesser threshold,twice pre-denoising and twice restoring are alternated.using the maximum-likelihood algorithm(PML),the wavelet packet threshold once denoising method and the twice denoising method are compared in experiment.Experimental results show that choosing lesser threshold and changing less energy distribution of degenerated image,the wavelet packet threshold twice denoising method can obtain better effect of image restoration than the once denoising method.