基于Contourlet变换的子带自适应图像去噪

Subband-adaptive image denoising based on contourlet transform

Contourlet变换是继小波变换之后的又一新变换。由于Contourlet变换的多尺度和多方向特性,能有效地捕获到自然图像中的轮廓,并对其进行稀疏表示.本文提出一种基于Contourlet变换子带自适应图像的新颖去噪算法。该算法核心是估计无噪期望信号的概率。即结合无噪子带系数的广义Laplacian模型和加性高斯白噪声的概率估计,分析每个子带信号概率为固定的情况。实验结果显示这种新的子带自适应图像去噪算法优于Bayesian wavelet shrinkage和ContourletHMT算法。

The contourlet transform is a new extension of the wavelet transform in two dimension. Because of its multiscale and directional properties, the contourlct transform can effectively capture the smooth contours that are the dominant features in natural images with only a small number of coefficients. We proposed a novel contourlet domain denoising method for subband - adaptive image denoising. The core of our approach is estimation of the probability that a given coefficient contains a significant noise - free component, which we call ＂signal of interest＂. In this respect we analyze case where the probability of signal presence is fixed per subband. All the probabilities are estimated assuming generalized Laplacian prior for noise - free subband data and additive white Gaussian noise. The experiment results show that the new subband - adaptive shrinkage denoising outperforms the Bayesian wavelet shrinkage and the cootourlet HMT.