基于Contourlet变换的图像消噪处理

文中提出了一种基于区域和Contourlet变换相结合的图像融合新方法，新方法利用小波Contourlet变换良好的多尺度性和多方向性特征，结合Contourlet变换的多聚焦图像融合算法，得到高频和低频图像。小波Contourlet由于变换缺乏平移不变性而产生图像失真，弱边缘具有几何信息，而噪声却不具备这个性质。为了在去噪声的同时更多地保留弱边缘信息，再次利用到循环平移的方法。实验结果表明，该文方法不仅在客观评价指标上优于小波变换法，而且从主观评价上来看，文中所提方法得到的图像更加清晰。

Contourlet Transformation-based Image Denoising Processing

This article proposes a new image fusion method based on the combination of region and contourlet transformation. This new method uses the good multi-criteria and multi-direction of the wavelet Contourlet transformation and combines the multi-focusing image fusion algorithm of Contourlet transformation. The wavelet Contourlet transformation, due to the lack of translation invariability has resulted in the image distortion, and the weak edge has the geometry information, while the noise actually has no such property. In order to retain the weak edge information while denoise, the circulation translation method is used again. The experimental result indicates that the method proposed in this article surpasses the wavelet transformation method in the objective evaluation index. And also, according to the subjective appraisal, the image obtained by the proposed method is much clearer.