|Table of Contents|

Color image denoising with block K-SVD dictionary learning

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
2016年05期
Page:
607-
Research Field:
Publishing date:

Info

Title:
Color image denoising with block K-SVD dictionary learning
Author(s):
Liu Xiaoman1Liu Yongmin2
1 Department of Mathematics,Southeast University,Nanjing 211189,China; 2 School of Mathematics and Statistics,Jiangsu Normal University,Xuzhou 221116,China
Keywords:
dictionary learning image segmentation color image denoising
PACS:
TP391
DOI:
10.14177/j.cnki.32-1397n.2016.40.05.017
Abstract:
The block K-singular value decomposition(K-SVD)dictionary learning algorithm for the color image denoising is contructed here.The noise is added to the color image firstly,and the image is divided into three grayscale images by the color channel,then the dictionary learning algorithms including the overcomplete discrete cosine transform(DCT)dictionary,the global trained dictionary and the adaptive dictionary are used to denoise grayscale images.With the image segmentation theory,the block K-SVD algorithm can be more effective operation for the dictionary learning.The experimental results show that,the DCT dictionary is suitable for the weak noise under the smaller weights,the global trained dictionary is suitable for the weak noise under the weight close to 1,and the adaptive dictionary is suitable for the weak noise under larger weights.

References:

[1] Donoho D L.Compressed sensing[J].IEEE Transactions on Information Theory,2006,52:1289-1306.
[2]Candes E.Compressive sampling[C]//Proceedings of International Congress of Mathematicians.Zurich,Switzerland:European Mathematical Society Publishing House,2006:1433-1452.
[3]Baraniuk R.Compressive sensing[J].IEEE Signal Processing Magazine,2007,24(4):118-121.
[4]刘亚新,赵瑞珍,胡绍海,等.用于压缩感知信号重建的正则化自适应匹配追踪算法[J].电子与信息学报,2010,32(11):2713-2717.
Liu Yaxin,Zhao Ruizhen,Hu Shaohai,et al.Regularized adaptive matching pursuit algorithm for signal reconstruction based on compressive sensing[J].Journal of Electronics and Information Technology,2010,32(11):2713-2717.
[5]Zhu Yonggui,Liu Xiaoman.A fast method for L1-L2 modeling for MR image compressive sensing[J].Journal of Inverse and Ill-Posed Problems,2015,23(3):211-218.
[6]Liu Xiaoman,Zhu Yonggui.A fast method for TV-L1-MRI image reconstruction in compressive sensing[J].Journal of Computational Information Systems,2014,10(2):1-9.
[7]刘梓,宋晓宁,於东军,等.基于多成分字典和稀疏表示的超分辨率重建算法[J].南京理工大学学报,2014,38(1):1-5.
Liu Zi,Song Xiaoning,Yu Dongjun,et al.Super-resolution reconstruction algorithm based on multi-component dictionary and sparse representation[J].Journal of Nanjing University of Science and Technology,2014,38(1):1-5.
[8]封晓强,何铁军.基于量子进化算法的交通图像稀疏分解[J].南京理工大学学报,2010,34(1):40-45.
Feng Xiaoqiang,He Tiejun.Sparse decomposition for traffic images using quantum-inspired evolutionary algorithms[J].Journal of Nanjing University of Science and Technology,2010,34(1):40-45.
[9]Aharon M,Elad M,Bruckstein A.K-SVD:An algorithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322.
[10]刘平,刘晓曼,朱永贵.基于K-SVD字典学习的核磁共振图像重建方法[J].中国传媒大学学报(自然科学版),2013,20(4):34-39.
Liu Ping,Liu Xiaoman,Zhu Yonggui.MR image reconstruction based on K-SVD dictionary learning[J].Journal of Communication University of China(Science and Technology),2013,20(4):34-39.
[11]Kreutz-Delgado K,Murray J F,Rao B D,et al.Dictionary learning algorithms for sparse representation[J].Neural Comp,2003,15(2):349-396.
[12]陈沅涛,徐蔚鸿,吴佳英,等.基于增量学习向量SVM方法的图像分割应用[J].南京理工大学学报,2014,38(1):6-11.
Chen Yuantao,Xu Weihong,Wu Jiaying,et al.Image segmentation application based on incremental learning vector SVM algorithm[J].Journal of Nanjing University of Science and Technology,2014,38(1):6-11.
[13]Mairal J,Elad M,Sapiro G.Sparse representation for color image restoration[J].IEEE Transactions on Image Processing,2008,17(1):53-69.
[14]Vogel C R.Computational methods for inverse problems[M].Philadelphia,US:SIAM,2002.
[15]姚敏.数字图像处理[M].北京:机械工业出版社,2012.

Memo

Memo:
-
Last Update: 2016-10-30