基于差分法的噪声检测和变分法的图像滤波算法

针对图像的椒盐噪声滤除算法中,在噪声检测阶段对噪声点的检测通常不够准确,在噪声恢复阶段,又缺乏对边缘信息的保护,文中提出了一种两步复原法,以用于复原被脉冲噪声破坏的模糊图像。算法将滤噪过程分为噪声检测和噪声恢复阶段。噪声检测过程中,在滑动窗口扩大当前的像素值和其他像素值之间的有序差异,来确定当前像素是否为噪声像素。而在噪声恢复过程中利用变分法,确保图像的边缘和细节。实验结果表明,文中所提检测、降噪方法在噪声密度较高的情况下,优于其他算法。

Noise Detection Based on the Difference Method and Image Filter Algorithm Based on the Variational Method

The salt and pepper noise filtering algorithm of the image is not accurate enough in the noise testing phase,and it can not protect the edge information in the recovery phase noise. In this paper,we propose a twophase approach to restore images corrupted by blur and impulse noise. The proposed filter has a two stage scheme：detecting noise and removing noise. First,an impulse noise detector is presented,by augmenting the ordered difference of the current pixel value with other pixels’ value in the sliding window to determine whether the current pixel is a noise pixel or not. Then,these noise pixels are restored using the variational method,which can preserve image edges and details. Experimental results show that the method proposed in this paper is superior to other methods at very high noise level.