基于小波域的图像噪声估计新方法

Estimating Image Noise Based on Region Segmentation in the Wavelet Domain

提出一种基于小波域区域分割的估计图像噪声的新方法。该方法利用图像的小波高频系数，在提出图像平滑区域的基础上，准确地估计图像高斯噪声的标准方差。由于考虑了图像的局部信息，因此该方法优于传统的估计方法。用于多幅实验图像的结果表明：在图像受噪声比较小或图像含高频信息较丰富时，该方法比传统方法更准确。

A new method is proposed to estimate image noise. The algorithm can improve traditional ones because it allows additional local information of the image (such as the identification of smooth or edge regions), and it recovers the variance of the noise in two steps. First, based on the distribution of wavelet high-frequency coefficients,the smooth regions in an image are extracted and a variance estimate sequence for these regions is yielded. In the second part of the algorithm, the value of the noise variance is determined from this variance estimate sequence. This paper applies the blind noise variance algorithm to some noisy images often employed in computer vision and image processing. Experimental results are provided which indicate that the novel method can provide better result than other traditional ones. These results are useful in image denoising.