摘要
图像中椒盐噪声的有效去除,取决于噪声检测和噪声灰度估测的准确性,但现有的滤波算法在噪声检测和噪声灰度估测上的准确性不高。因此,提出了基于图像纹理特征的决策滤波算法。算法根据椒盐噪声的灰度最值特征和独立性,以及图像纹理的特征进行噪声检测,将噪声与信号像素准确地区分开。算法根据纹理中像素灰度的平滑变化特征,将邻域中的信号像素进行分组,然后基于相关性与正态概率分布的意义,取与邻域均值最接近的分组的中值作为噪声像素的估测值。实验的结果证明,所提出的算法检测噪声更加准确,其去噪结果对应的峰值信噪比(PSNR)比现有的算法平均提高1.9 dB以上,图像增强因子(IEF)比现有的算法平均提高119以上。因此,相对于现有的算法,所提出的算法在去噪性能上具有显著的优越性。
The effectiveness of salt and pepper noise removal lies on the noise detection and the intensity estimation of noisy pixel,but existing filters show low performance for them.In view of this problem,a decision filter based on image texture features is proposed.The proposed method performs noise detection by taking full advantage of the characteristics of salt and pepper noise,i.e.,the noise takes extreme intensity values and is independent on noise free pixels,as well as the texture features of image,thus,correctly discriminates the noisy pixels from the noise free pixels.Based on the varied intensity of image texture,the proposed method groups the neighbor noise free pixels,in the light of correlation and significance of normal probability distribution,it takes the median of the group,which is closest to the mean of neighbor noise free pixels;as the intensity of noisy pixel.The experimental results show that,the proposed method can perform a more accurate noise detection compared with the existing filters,it resulted that PSNR increases averagely by more than 1.9 dB,IEF increases averagely by more than 119.Thus,it can conclude that the denoising performance of the proposed method is superior to the existing filters.
作者
陈家益
战荫伟
曹会英
董梦艺
Chen Jiayi;Zhan Yinwei;Cao Huiying;Dong Mengyi(School of Information Engineering,Guangdong Medical University,Zhanjiang,524023,China;School of Computer Science and Technology,Guangdong University of Technology,Guangzhou 510006,China;Second Clinical Medical College,Southern Medical University,Guangzhou 510515,China)
出处
《电子测量与仪器学报》
CSCD
北大核心
2019年第3期126-135,共10页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(61170320)
广东省自然科学基金(2015A030310178)
广东省科技计划(2017B010110015)
广州市科技计划(201604016034)
广东省医学科研基金(B2018190)
湛江市科技攻关计划(2017B01142)
广东医科大学科研基金(GDMUM201815
GDMUM201827)资助项目
关键词
图像去噪
噪声检测
中值滤波
灰度最值
纹理特征
决策滤波
image denoising
noise detection
median filter
extreme intensity values
texture feature
decision filter
