图像脉冲噪声的模糊滤波算法研究

Study on Fuzzy Filter Algorithm for Image Impulse Noise

论述了基于模糊规则的脉冲噪声滤波器. 该滤波器由模糊脉冲噪声检测器、噪声消除器与模糊结合器构成. 模糊脉冲噪声检测器用窗口内的中值与邻近像素信息来检测脉冲噪声，而脉冲消除器用最小最大值算法计算噪声像素的估计值. 与传统的脉冲噪声滤波器相比，所设计的新滤波器具有良好的脉冲噪声抑制与图像细节边缘保护的性能.

In this paper, a new filter is presented which adopts fuzzy rules. The filter consists of a fuzzy impulse noise detector, impulse noise canceller and fuzzy jointer. The fuzzy impulse noise detector uses the median value and neighborhood information in the window to detects impulse noise. The impulse noise canceller uses minimum-maximum estimate method. The fuzzy filter is better at preserving fine details and suppressing impulse noise than other impulse noise filters.