中文摘要：

提出了一种基于DS证据理论的滤波方法，用于去除两种类型的脉冲噪声，这种新的滤波方法由检测和滤波两部分组成。首先获得证据，并用窗中的信息定义集合函数；然后用决定规则来判定噪声是否存在。最后在滤波过程中，对检测到的噪声点用本文中的滤波方法进行处理，而好的象素点保持不变，这就避免了对非噪声象素点的破坏。从实验结果中可以看到，本文的方法对固定值和随机值的脉冲噪声都能有效抑制，并且可以保留图像细节。

英文摘要：

This paper proposes a filtering method based on DS evidence theory, which is used for removing two types of impulse noise. This new method is composed of impulse detection and noise filtering. We should get evidence, and define mass function by local information. Using a rule determine whether noises exist. During the filtering process, the noisy pixel is dealt with method proposed in this paper. The pixels uncorrupted remain unchanged. So those pixels uncorrupted are protected well. From the result, we can see that the method proposed in this paper is effective to two types of impulse noise, and can retain details in image at the same time.