我们提出一个新的方法，方法噪声，用来估计和比较数字图像去噪方法的表现。我们首先对一个广泛的去噪算法类计算和分析方法噪声，称之为局部平滑滤波。其次，我们提出一种新的基于图像中所有像素的非局部平均的算法，非局部平均（NL-means）。最后，我们展示几组实验比较非局部平均算法和局部平滑滤波算法。-We propose a new method for noise method used to estimate the performance of digital image denoising methods and compare. We first noise, a denoising algorithm calculation and analysis method called partial smoothing filter. Secondly, we have proposed a new algorithm based on all the pixels in the image non-local average, the non-local average (NL-means). Finally, we show several groups of experiments comparing non-local averaging algorithm and the local smoothing filter algorithm.