结合波原子和Cycle Spinning的图像去噪

Image denoising based on wave atoms and cycle spinning

结合波原子变换和Cycle Spinning的优点,提出一种新的图像去噪算法.由于波原子变换缺乏平移不变性,直接进行系数阈值去噪会在去噪图像边缘产生伪吉普斯现象,导致图像的失真.该算法引入Cycle Spinning技术有效抑制这种视觉失真,对原始含噪图像进行波原子硬阈值去噪.实验结果表明,与单一波原子变换、小波Cycle Spinning方法相比,新算法能够在去除噪声的同时保留边缘,有效抑制了传统去噪方法的伪吉普斯现象,视觉效果也能得到较好的改善.对强噪声级的图像,这种优势更为明显.

A new method for image denoising is presented,which colligates the strongpoint of the wave atoms transform and Cycle Spinning.Due to lack of translation invariance of the wave atoms transform,image denoising by coefficient thresholding would lead to Pseudo-Gibbs phenomena.Cycle Spinning is employed to avoid the artifacts.Experimental results show that the method can remove noisy and remaining edges,while Pseudo-Gibbs phenomena are controled efficiently,and can get a better visual effect and PSNR gains compared with the methods like simplex wave atoms or wavelet denoising using Cycle Spinning.And in heavy background noise,this advantage is significant.