基于排序跳变点的脉冲噪声检测与滤除算法

提出了一种简单的判断脉冲噪声点的算法。该算法先将滤波窗口内的像素值按照大小排序,然后找到排序后像素值的两个最大跳变点,如果此跳变点发生在靠近最大值或最小值处,并且跳变幅度超过文中实验得出的域值,就可以根据窗口中心像素值的位置,即它在两个跳变点所确定的范围之内还是之外,将其判为信号点或噪声点,最后用判断出的信号点对噪声点进行恢复。实验结果表明,该算法能够几乎准确地判断出噪声点与信号点,从而达到了更好的去噪效果。

Impulse Noise Detection and Filtering Algorithm Based on Leap Points of Rank-ordered Pixels

This paper proposes a novel algorithm for detecting noise pixels in an image corrupted by impulse noise. Firstly,the pixels in the filter-window are rank-ordered according to their magnitudes,Then the two most significant leap points are found.lf they are close to the minimum or maximum value and exceed the extent concluded by the author, we can judge the center pixel as noise pixel if its magnitude is outside of the range confined by the leap points.At last,we restore the noise pixels on the basis of the signal pixels,The experimental results indicate that our algorithm can detect pixels contaminated by impulse noise accurately,and achieves better de-noising effects.