基于邻域相关的自适应中值滤波算法

Adaptive Median Filter Based on Neighborhood Correlation

为改善图像效果,在对现有的椒盐噪声自适应中值滤波算法分析的基础上,提出了基于邻域相关的自适应中值偏移滤波算法。算法对自适应中值滤波算法的处理过程进行改进。当滤波窗口大于设定的最大值时,根据像素点邻域内的像素点灰度值分布在不同的区间,将像素点灰度值的复原分为偏大或偏小两种情况。受污染像素点的滤波输出值为偏移的中值,偏移的距离根据改进的自适应中值滤波算法运行时滤波窗口的大小来决定。从仿真结果可见:该算法滤波后的图像具有较高的峰值信噪比,对图像细节的保留能力和抑制椒盐噪声的能力都有一定的提高。

To improve image effect,a new improved adaptive median offset filtering based on the statistic of the neighborhood pixels is put forwarded,through analyzing existing methods of adaptive median filtering on salt and pepper noise.The proposed algorithm improves process of adaptive median filtering.Based on the distribution of gray value of neighborhood pixels,when the filter window size is larger than the setting maximum,the recovery of the gray value of pixels is divided into large or small case.The gray value of contaminated pixel is reconstructed by offset median.The offset distance is determined by size of filter window of improved adaptive median filtering.Simulation results show that this proposed method has a better improvement on preserving the details of image and removing salt and pepper noise.The restored image can obtain superior peak signal to noise ratio.