基于最值中滤波的彩色图像噪声滤波器

An Extremum Median Fidian Filter for Impulsive Noise Removal In Color Images

构建了一种新的消除彩色图象中脉冲噪声的滤波法。该滤波首先根据极值点(最大或最小)的判别标准将全部像素分为两类(信号S与噪声N),再根据空间相关性将噪声点用该点邻域的中值取代,然后进行简单的局部纹理分析而自动选择使用标量中值滤波或者使用矢量中值滤波。实验表明,文中提出的方法计算量少,性能优良,图象信噪比高。

This praper proposes a new filtering approach for removing impulsie in color images. The filtre, according tothe decisterion, is given in the ofllowing: if a pixel valie is the extremum(max.or min.)of its neighborhood, it is a noise pixel;else, it is a signal pixel; then noise pixels are replaced with the median value of their neighborhood in the input image. Next,the proposed approach automtically chooses to use either a extrmum scalar median filter or a vector median filter according tosimple local texture analysis. Experiments show that the proposed method provides comparable performance with much lesscomputation requirement than a standard vector median filter. And, the extremum median filter have a eminent capability,theproduce images' SNR is high.