灰度极限脉冲耦合神经网络混合噪声滤波

摘 要:针对图像中同时存在椒盐噪声和高斯噪声,提出一种基于灰度极限和脉冲耦合神经网络(PCNN)滤除混合噪声的新方法。首先,根据灰度极值定位出椒盐噪声点;其次,在滤波窗口中对椒盐噪声点进行均值滤波;然后,利用PCNN赋时矩阵定位出高斯噪声点;最后,自适应调整可变灰度步长,选择不同滤波方法滤除高斯噪声。实验结果表明提出的算法较常见的混合噪声滤波方法在主观滤波效果和客观评价指标峰值信噪比(PSNR)及信噪比改善因子(ISNR)两方面均有明显的优势。

关键词:椒盐噪声;高斯噪声;灰度极限;脉冲耦合神经网络;均值滤波;可变步长

中图分类号: TP391.4 文献标志码:A

Mixed noise filtering via limited grayscale pulse coupled neural network

CHENG Yuanyuan1, LI Haiyan1*, CHEN Haitao2, SHI Xinling1

(

1.School of Information Science and Engineering, Yunnan University, Kunming Yunnan 650091, China;

2.Ultrasound Department, The Third Affiliated Hospital, Medical University of Yunnan, Kunming Yunnan 650118, China

)

Abstract:

A new method of filtering mixed noise based on limited grayscale and Pulse Coupled Neural Network (PCNN) was proposed for an image contaminated by salt and pepper noise and Gaussian noise. First, salt and pepper noise was identified according to the limited grayscale in a detecting window. Then the noise was filtered via mean filter in a filtering window. Subsequently, Gaussian noise was identified by using the time matrix of PCNN. Finally the Gaussian noise was filtered by some different filters based on variable step. The experimental results show that the proposed method has more advantages not only in filtering effects but also in objective evaluation indexes of Peak SignaltoNoise Ratio (PSNR) and Improved SignaltoNoise Ratio (ISNR) compared to some traditional methods.

Key words:

salt and pepper noise; Gaussian noise; limited grayscale; Pulse Coupled Neural Network (PCNN); mean filtering; variable step

0 引言

在实际的图像通信系统和图像处理系统中,由于外界的干扰和破坏,图像在获取、传输过程中不可避免地受到噪声污染,这些噪声对图像的后续处理产生很大影响,所以滤波一直是图像研究领域的热点。图像中混合噪声常常包含椒盐噪声和高斯噪声。传统中值滤波和均值滤波对混合噪声的去噪效果较差,现有研究较多是针对其中一种噪声进行滤波[1-7],而针对混合噪声的研究[8-10]较少。文献[8]结合多级中值滤波、均值滤波和脉冲耦合神经网络(Pulse Coupled Neural Network, PCNN)对混合噪声滤波,文献[9]提出自适应中值滤波和自适应压缩加权均值混合噪声滤波方法,文献[10]提出利用局部阈值将高斯噪声和椒盐噪声区分之后单独对其滤波,这些方法对图像的混合噪声滤波效果均优于传统滤波方法,但在噪声强度较大时,主观去噪效果与客观评价指标均明显下降。

因此,本文提出一种灰度极限脉冲耦合神经网络(Limited GrayscalePulse Coupled Neural Network, LGPCNN)的新方法来滤除图像混合噪声。首先,根据椒盐噪声极值特性,利用图像灰度极值来判断椒盐噪声点;然后,利用均值滤波器滤除椒盐噪声,这样有效地保留了图像的细节信息;最后,利用PCNN赋时矩阵,选择不同滤波方法滤除高斯噪声。实验结果证实，本文算法对混合噪声的滤波效果显著,优于传统滤波方法和文献[8-10]提出的滤波算法。

1 滤波算法

1.1 椒盐噪声滤除

图像含有的椒盐噪声表现为一些极暗或极亮点,趋于极值特性,所以提出算法先对含噪图像的灰度值进行极值检测,把检测出的极值点作为椒盐噪声点,再利用均值滤波滤除椒盐噪声。

设图像中某一点为Ixy,在以Ixy 为中心、大小为5×5的检测窗口中,判定图像灰度最大值Imax和最小值Imin的像素为椒盐噪声点,介于最大值和最小值之间的像素点为非椒盐噪声点。若判断出Igk是一个椒盐噪声点,则以Igk为滤波窗口的中心,噪声点的灰度值变为滤波窗口中所有非椒盐噪声点灰度值的均值。若在滤波窗口中不存在非椒盐噪声点,则扩大滤波窗口,重复检测,直到窗口中含有非椒盐噪声点。本文经过大量实验,最终选定滤波窗口为5×5。由于该种方法只利用窗口中的非椒盐噪声点进行噪声滤波处理,排除了椒盐噪声点的影响,所以对椒盐噪声点的滤除达到较好的效果。