模糊加权中值滤波器

本文介绍了一种模糊加权中值滤波器，该滤波器由模糊布尔函数和滤波加权确定。本文用S型函数逼近模糊布尔函数。此外，用模糊理论领域中使用的S型函数逼近所滤波的加权。模糊加权中值滤波器只由4个参数确定。所提出的滤波在均方误差准则下能够由最小均方算法导出。图像复原的实验结果表明，本文介绍的模糊加权中值滤波方法既能去除脉冲噪声和平滑高斯噪声，又能同时有效地保持边缘和图像细节，漠糊加权中值滤波器明显优于加权中值滤波器，也优于Wiener滤波器。

Fuzzy Weighted Median Filters

In this paper a fuzzy weighted median(FWM)filter is proposed. The FWM filter is defined by the Boolean function and filter weights. We approximate the fuzzy Boolean function by the S-type function. Furthermore, we approximate weights of the filter by the S-type function which is used in the field of the fuzzy theory. The FWM filter is defined by only four parameters. The proposed filter can be derived by least mean square(LMS)algorithm under the mean square error criterion. The experimental results in image restoration show that the proposed FWM filters are able to remove the impulse noise and smooth Gaussian noise as well as efficiently preserve edges and image details. Our conclusion is that the FWM filters have obvious superiorities over not WM filters but also Wiener filter.