基于Contourlet变换的带噪图像增强方法

为了解决带噪图像增强中抑制噪声和增强边缘细节的矛盾,提出一种基于Contourlet变换的带噪图像增强方法。Contourlet变换具有多分辨率、局部定位性、多方向、各向异性等特点,比小波变换更能有效地捕获图像中的高维奇异性。根据这一特点,本文综合考虑变换后系数尺度间和尺度内的依赖性,先在Contourlet变换域中设置自适应阈值抑制噪声;在此基础上,应用广义非线性增益函数来提高较弱细节的局部对比度。实验结果表明,该算法较好的抑制噪声,增强图像细节对比度。

Noisy Image Enhancement Algorithm Based on the Contourlet Transform

A new image enhancement method aimed at optimizing contrast of image features while minimizing image noise is pro- posed in the paper based on the Contourlet transform.Contourlet transform is is multiresolutional,localized,multidirectional and aniso- tropic,so it is more effectively capture high dimensional singularity.Because of all these characteristics,and considering of the intrasub- band and interscale dependencies,firstly an adaptive shrinkage denoising is accomplished.Then the generalized nonlinear adaptive agin function is used to enhance the features with low contrast while protecting the strong contrast features from over enhancing.Results of ex- periments show that this method enhance the contrast of the image while avoiding over-enhancement of noise.