图像快速自适应加权中值滤波

Fast Adaptive Weighted Median Filter for Image

中值滤波由于可对随机噪声起到良好的去除效果因而在信号恢复中得到了较为广泛的应用。本文提出了一种新的快速自适应加权中值滤波算法以提高中值滤波算法的性能。该算法通过噪声检测确定图像被污染的程度,再根据污染情况选取不同的加权值。为提高中值滤波的速度,作者基于数组分布范围有限这一特征提出一种多级分阶的统计直方图中值求解方法。实验证明该算法在抑制噪声的同时也能较好地保护细节信息,并且能有效地提高标准中值滤波的速度。

Due to its effectiveness for removing impulse noise ,median filtering has long been a popular tool for signal restoration ,In this work , a new fast adaptive weighted median filter（FAWMF） is proposed for improving the performance of median-based filters. In this algorilhm, noise detection is used to determine impulse noise and the degree of the noise interference .Based on the degree of noise interference, weighted value could be choosed,To improve the speed of median filter, we present a new fast median filter which is based on statistical histogram and multilevel staged search according to the character that the array distribution scope is limited. Experimental resuhs show that the proposed filter not only can provide better performance of suppressing impulse but can preserve more detail features, and it effectively improves the speed of standard median fiher method.