中文摘要：

为了有效地抑制由椒盐噪声和高斯噪声组成的混合噪声，提出一种基于灰色关联系数的混合噪声滤波算法。检测出椒盐候选噪声点，采用动态滤波窗口对椒盐候选噪声点进行分类处理，利用均值滤波后的图像信息有选择地修正受高斯噪声影响较重像素点的灰度值，对其进行高斯滤波处理。实验结果表明，该算法能有效地滤除混合噪声，提高图像的清晰度，为研究灰色关联理论滤除混合噪声提供一种有效的途径。

英文摘要：

To effectively control the mixed noise composed of salt and pepper noise and the Gaussian noise, an algorithm tor mixed noise filtering is proposed based on grey relational coefficient. The candidate noise point caused by salt and pepper noise is detected. Secondly, the candidate noise point is classified by using dynamic filter window. Then, selectively modify the grey value of the pixel point which heavily affected by Gaussian noise using the image after mean filtering. Finally, the Gaussian noise is filtered out. The experimental results show that the algorithm can effectively filter out mixed noise, enhance image clarity, and provides an effective way to filter mixed noises with grey relational theory.