期刊文献+

椒盐噪声图像的非局部平均滤波算法 被引量:1

A nonlocal means filter for images with salt-and-pepper noise
下载PDF
导出
摘要 针对非局部平均(NLM)方法对椒盐噪声图像滤波效果较差的问题,通过引入噪声检测结果扩展NLM方法去除图像中椒盐噪声。在噪声检测阶段,利用图像的两个极值Lmin和Lmax把图像像素点分为非噪声点和噪声点。在滤波阶段,非噪声点的灰度值保持不变。对于噪声点,如果以该噪声点为中心的自适应滤波窗口内均为噪声点,则认为该噪声点位于图像自身灰度值为Lmin或Lmax的区域内,使用两个极值的统计结果进行恢复。否则,采用改进的NLM方法滤除噪声。构造联合噪声检测模板避免噪声点对相似权计算的干扰,噪声点的恢复值由非噪声点的灰度值加权平均得到。此外,采用迭代滤波策略对高密度噪声图像噪声点进行恢复。相关去噪实验结果证实了算法去噪的有效性,不足之处是算法的时间复杂度较高。 According to the fact that the nonlocal means (NLM) method cannot adequately remove noise from the images corrupted by salt-and-pepper noise, we extend the NLM to remove salt-and-pep- per noise by introducing the noise detection results. At the noise detection stage, we divide pixels into two categories: noisy and noise-free pixels, depending on two extreme values Lm^n and L At the filte- ring stage, noise-free pixels remain unchanged, while for each noise pixel, if the adaptive filtering win- dow does not contain any noise-free pixel, we regard the current noise pixel located in image uniform re- gions composed of noise-free pixels with the same gray value Lmin or L And then the calculated statis- tics is used as the restored value. Otherwise, we employ the improved NLM filter for noise removal. The joint noise detection mask in the proposed method can avoid the influence of noise pixels on calculat- ing similar weights in the presence of noise pixels, and only noise-free pixels are used for the weighted average. In addition, the iterative filtering scheme is used to remove noise of high-density. Experimental results demonstrate the effectiveness of the proposed filter even though its computational complexity is still high.
出处 《计算机工程与科学》 CSCD 北大核心 2017年第6期1133-1148,共16页 Computer Engineering & Science
基金 国家自然科学基金(61471004) 安徽理工大学青年教师科学研究基金(QN201322) 安徽理工大学博 硕士基金(ZX942)
关键词 图像去噪 椒盐噪声 非局部平均 相似权 迭代滤波 image denoising salt-and-pepper noise nonlocal means similarity weight iterative filtering
  • 相关文献

参考文献2

二级参考文献21

  • 1司文武,钱沄涛.一种基于谱聚类的半监督聚类方法[J].计算机应用,2005,25(6):1347-1349. 被引量:11
  • 2李恒建,尹忠科,王建英.基于量子遗传优化算法的图像稀疏分解[J].西南交通大学学报,2007,42(1):19-23. 被引量:21
  • 3Pascal V and Yoshua B. Kernel matching pursuit[J]. MachineLearning, 2002, 48(1): 165-187.
  • 4Nguyen H V and Nasser M N. Design of non-linear kerneldictionaries for object recognition[J]. IEEE Transcations onImage Processing, 2013, 22(12): 5123-5135.
  • 5Ramin P, Hossein N Z, and Louis T. Auditory-inspired sparserepresentation of audio signals [J]. Speech Communication,2011,53(5): 643-657.
  • 6Adelino R and Sliva F D. Atomic decomposition withevolutionary pursuit [J]. Digitals Signal Processing' 2003,13(2): 317-337.
  • 7Kubo Y, Watanabe S, Nakamura A, et al" Basic vectororthogonalization for an improved kernel gradient matchingpursuit method[C]. 2012 IEEE International Conference onAcoustics, Speech and Signal Processing (ICASSP), Kyoto,Japan, 2012: 1909-1912.
  • 8Jiao L C and Li Q. Kernel matching pursuit classifierensemble [J]. Pattern Recognition, 2006,39(4): 587-594.
  • 9Zhao S L, Wu P, and Liu Y P. An online kernel learningalgorithm based on orthogonal matching pursuit [J], Journalof Software, 2012,7(9): 2076-2082.
  • 10Frey B J and Dueck D. Clustering by passing messagesbetween data points[J]. Science, 2007, 315(5814): 972-976.

共引文献10

同被引文献11

引证文献1

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部