一种基于两阶段的脉冲噪声滤除算法

A two-stage method for impulse noise reduction

本文提出了一种基于两阶段的脉冲噪声滤除方法。在算法的第一阶段,提出利用列队排序检测器(ROD)来检测图像中所有可能的脉冲噪声点。在第二阶段,对所有的噪声候选点进行自适应中值滤波,滤波窗口的尺寸大小是根据噪声密度自适应调整的。该算法能对图像的边界以及非噪声点进行保护。实验表明,本文算法在滤除脉冲噪声的同时可以有效地保护图像细节,尤其是在噪声密度非常大的情况下。

We propose a two-stage method for removing impulse noise. In the first stage,we use Rank Order Detector （ROD） to identify pixels which are likely to be corrupted by impulse noise. In the second stage,these noise candidates are restored using an adaptive median-based method which allows edges and noise-free pixels to be preserved. Simulation results indicate that the proposed method is significantly better than those using other impulse noise reduction filters.