基于Contourlet变换尺度相关多阈值降噪算法

该文研究了多方向多尺度几何分析工具轮廓波Contourlet变换，在总结了Contourlet变换稀疏尺度关系的基础上，提出了一种基于Contourlet变换多阈值尺度相关降噪算法，利用Contourlet变换多分辨率以及优越的方向性进行阈值降噪，根据Contourlet变换后各个尺度间的相关特性及噪声强度设定不同的阈值，由于Contourlet变换比小波变换更稀疏的表示方式，更好地表示图像的细节，在图像降噪的过程中能更好地保护图像的纹理以及边缘，使得降噪后的图像在细节方面更突出。

Multi-threshold Denoising Algorithm Based on Scale Correlation of Contourlet Transform

Researches the multi-scale geometry analysis tool -Contourlet, and proposes a denoising algorithm taked advantage of directional stability and anisotropy of Contourlet transform. The algorithm setting different thresholds based on the scale corre-lation of Contourlet transform and the noise intensity. As Contourlet transform is good at representing the image edge features, it can bettere protect the edges and texture of the image, and improve the visual effectly.