基于Contourlet变换在图像去噪中的应用

A method for image denoising based on Contourlet transform

Contourlet变换是小波变换的新发展,能更有效地捕捉图像中的几何结构。本文提出一种基于Contourlet变换的图像去噪方法。该方法通过对阈值去噪中的统一阈值和阈值函数进行分析,构造出一个新的阈值函数,新阈值函数在一定程度上改进了统一阈值"过扼杀"Con-tourlet系数的缺点,同时也使硬阈值去噪中出现的伪吉布斯现象得到解决。实验结果表明,本文采用的方法提高了去噪后图像的PSNR值,同时有效地保留了图像纹理信息,使视觉效果更好。

The Contourlet transform is a new development of wavelet transformation, it can be more effective in capturing geometric structure of images, a method for image denoising based on the Contourlet transform is proposed in this paper. Through the analysis of uniform threshold and threshold functions in image denoising, a new threshold function is constructed, the new threshold function deals with the problem of snuffing out the Contourlet coefficients to exceed, and it also can solve the problem of pseudo-Gibbs phenomena in hard-thrashold denoising. The experimental results show that our method can get higher PSNR value and better visual effect compared with other methods.