基于球坐标系的非线性小波变换去噪算法

Nonlinear Wavelet De-Noising Algorithms Based on Spherical Coordinates

提出一种新的非线性小波变换去噪方法 .该方法把线性小波变换后得到的 3个高频分量通过球坐标变换后 ,得到球坐标下 3个分量 ,然后在球坐标中根据图像统计特性设定的收缩因子处理小波收缩去噪 .结果表明 ,该方法比直接进行小波收缩的结果有更好的峰值信噪比 .

This paper presented a new method based on nonlinear wavelet transform for image denoising. Three high-frequency parts from wavelet transforms are turned into spherical coordinate, in which it applies the method for image denoising by the wavelet shrinkage parameters based on statistical characteristics. The results of experiment prove a better PSNR.