一种基于脉冲噪声检测的图像均值滤波方法

论文提出了一种针对图像脉冲噪声进行检测,并根据检测结果利用均值滤波方法滤除脉冲噪声的方法。该方法首先采用串行方式,对含有脉冲噪声的图像进行逐点检测,其中判断噪声点的阈值可自适应地调整。采用窗口长度自适应调整和选择性取样的均值滤波方法,对检测到的噪声点进行逐点滤除。该方法既可有效地滤除脉冲噪声,又可以较好地保持图像边缘细节,对图像的后续处理有很好的价值。论文最后给出了实验滤波结果,说明了该方法的有效性。

A Image Mean Filtering Algorithm Based on Impulse Noise Detection

The paper proposes an algorithm that removes the impulse noise in an image by mean filtering on the basis of the impulse noise detection result.It first detects impulse noises in an image in a serial mode,and the threshold,which is used to estimate noise pixel,can be modified adaptively.The mean filtering,whose window length can also be modified adaptively and pattern is selective,is used to remove the noises step by step.The algorithm can effectively re-move the impulse noises and remains the image edge detail at the same time.It is very worthy of the latter image pro-cessing.The filtering result by experiments is given and it testifies the effectiveness of the algorithm.