[1]汤中民,唐贵进,刘小花,等.一种新的字典更新和原子优化的图像去噪算法[J].计算机技术与发展,2019,29(04):33-37.[doi:10. 3969 / j. issn. 1673-629X. 2019. 04. 007]
　TANG Zhong-min,TANG Gui-jin,LIU Xiao-hua,et al.A New Image Denoising Algorithm Based on Dictionary Updating and Atom Optimization[J].,2019,29(04):33-37.[doi:10. 3969 / j. issn. 1673-629X. 2019. 04. 007]
点击复制

一种新的字典更新和原子优化的图像去噪算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年04期
页码:
33-37
栏目:
智能、算法、系统工程
出版日期:
2019-04-10

文章信息/Info

Title:
A New Image Denoising Algorithm Based on Dictionary Updating and Atom Optimization
文章编号:
1673-629X(2019)04-0033-05
作者:
汤中民; 唐贵进; 刘小花; 崔子冠; 刘峰
南京邮电大学 江苏省图像处理与图像通信重点实验室,江苏 南京 210003
Author(s):
TANG Zhong-min; TANG Gui-jin; LIU Xiao-hua; CUI Zi-guan; LIU Feng
Key Lab on Image Processing and Image Communication of Jiangsu Province,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
关键词:
字典学习; 稀疏表示; 顺序更新; 字典优化; 图像去噪
Keywords:
dictionary learning; sparse representation; sequence updating; dictionary optimization; image denoising
分类号:
TP391. 41
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 04. 007
摘要:
经典的K-奇异值分解(K-SVD)算法通过字典对图像进行稀疏表示,在去噪的同时保持了原图像的有效信息。但是在基于噪声图像字典学习所得到的学习字典中通常含有大量的噪声信息,这也使得恢复出的图像仍然含有许多噪声,特别是在强噪声下,该算法性能表现较差。鉴于K-SVD算法的局限性,提出了一种新的基于字典更新和字典原子优化的图像去噪算法。首先利用一种加权的顺序字典学习(SDL)方法替代K-SVD算法,在字典更新阶段添加稀疏约束,这样能够得到更为稀疏的表示图像的字典;然后自适应地根据图像的结构复杂度和噪声强度进行字典原子检测并删除噪声原子;最后利用优化后的字典重构图像。实验结果表明,该算法与经典K-SVD、SDL等去噪算法相比,能够取得更好的去噪效果。
Abstract:
The classical K-singular value decomposition (K-SVD) algorithm sparsely expresses the image through the dictionary,and retains the original image’s effective information while denoising. However,the learning dictionary based on the noise image dictionary usually contains a large amount of noise information,which also makes the recovered image still contain many noises. Especially under strong noise,the performance of the algorithm is poor. In view of the limitations of the K-SVD algorithm,we propose a new image denoising algorithm based on dictionary updating and dictionary atom optimization. Firstly, a weighted sequential dictionary learning (SDL) method is used instead of the K-SVD algorithm to add sparse constraints in the dictionary update phase,so that a more sparse representation of the image dictionary can be obtained. Second,the dictionary noise atom is detected adaptively according to the structural complexity and noise intensity of the image and removed. Finally,the image can be reconstructed by using the optimized dictionary. The experiment shows that this algorithm can achieve better denoising results compared with the classical K-SVD,SDL and other denoising algorithms.

相似文献/References:

[1]朱伟冬 胡剑凌.基于马氏距离的稀疏表示分类算法[J].计算机技术与发展,2011,(11):27.
　ZHU Wei-dong,HU Jian-ling.Sparse Representation Classification Algorithm Based on Mahalanobis Distance[J].,2011,(04):27.
[2]王韦刚 庄伟胤.基于NIOS Ⅱ的图像压缩感知[J].计算机技术与发展,2012,(04):12.
　WANG Wei-gang,ZHUANG Wei-yin.Compressed Sensing of Image Based on NIOS Ⅱ[J].,2012,(04):12.
[3]葛广重,杨敏.基于稀疏表示的单幅图像超分辨率重建[J].计算机技术与发展,2013,(09):113.
　GE Guang-zhong,YANG Min.Single Image Super-resolution Reconstruction Based on Sparse Representation[J].,2013,(04):113.
[4]赵海峰,于雪敏,邹际祥,等.基于L1范数主成分分析的颅脑图像恢复[J].计算机技术与发展,2014,24(01):231.
　ZHAO Hai-feng[],YU Xue-min[],ZOU Ji-xiang[],et al.Cerebral Image Recovery Based on L1-norm Principal Component Analysis[J].,2014,24(04):231.
[5]陈静,邱晓晖,孙娜. 基于二维Gabor小波与SPP算法的人脸识别研究[J].计算机技术与发展,2014,24(11):110.
　CHEN Jing,QIU Xiao-hui,SUN Na. Research on Face Recognition Based on 2 D Gabor Wavelet and SPP Algorithm[J].,2014,24(04):110.
[6]姚刚,杨敏. 稀疏子空间聚类的惩罚参数自调整交替方向法[J].计算机技术与发展,2014,24(11):131.
　YAO Gang,YANG Min. Alternating Direction Method of Self-adjusting Penalty Parameters of Sparse Subspace Clustering[J].,2014,24(04):131.
[7]徐静妹,李 雷.基于稀疏表示和支持向量机的人脸识别算法[J].计算机技术与发展,2018,28(02):59.[doi:10．3969/j．issn．1673－629X．2018．02．014]
　XU Jingmei,LI Lei.A Face Recognition Algorithm Based on Sparse Ｒepresentation and Support Vector Machine[J].,2018,28(04):59.[doi:10．3969/j．issn．1673－629X．2018．02．014]
[8]钱阳,李雷. 一种基于新型KPCA算法的视频压缩感知算法[J].计算机技术与发展,2015,25(10):101.
　QIAN Yang,LI Lei. A Video Compressed Sensing Algorithm Based on Novel KPCA[J].,2015,25(04):101.
[9]于云,周伟栋. 基于稀疏表示的鲁棒性说话人识别系统[J].计算机技术与发展,2015,25(12):41.
　YU Yun,ZHOU Wei-dong. Robust Speaker Recognition System Based on Sparse Representation[J].,2015,25(04):41.
[10]谢文浩,翟素兰. 基于加权稀疏近邻表示的人脸识别[J].计算机技术与发展,2016,26(02):22.
　XIE Wen-hao,ZHAI Su-lan. Face Recognition Based on Weighted Sparse Neighbor Representation[J].,2016,26(04):22.
[11]余琨,荆晓远,吴飞,等. 基于竞争聚集的K-SVD字典学习算法[J].计算机技术与发展,2015,25(11):44.
　YU Kun,JING Xiao-yuan,WU Fei,et al. K-SVD Dictionary Learning Algorithm Based on Competitive Agglomeration[J].,2015,25(04):44.
[12]李欣,崔子冠,陈杰,等. 基于局部回归和自相似性的图像超分辨率重建[J].计算机技术与发展,2016,26(10):17.
　LI Xin,CUI Zi-guan,CHEN Jie,et al. Image Super-resolution Reconstruction Based on Local Regression and Self-similarity[J].,2016,26(04):17.
[13]钱阳,李雷. 基于新型鲁棒字典学习的视频帧稀疏表示[J].计算机技术与发展,2017,27(02):37.
　QIAN Yang,LI Lei. Sparse Representation of Video Frame Based on Novel Robust Dictionary Learning[J].,2017,27(04):37.
[14]钱阳,李雷,袁安安. 基于自适应K-SVD字典的视频帧稀疏重建算法[J].计算机技术与发展,2017,27(06):36.
　QIAN Yang,LI Lei,YUAN An-an. An Adaptive K-SVD Dictionary Learning Algorithm for Video Frame Sparse Reconstruction[J].,2017,27(04):36.
[15]陈春林,张礼,刘学军. 针对SAR图像的树形稀疏表示结构识别算法研究[J].计算机技术与发展,2017,27(08):20.
　CHEN Chun-lin,ZHANG Li,LIU Xue-jun. Investigation on Identification Algorithm of Tree-structure Sparse Representation for SAR Target[J].,2017,27(04):20.
[16]沈怡灵,赵明哲,李强懿,等.基于稀疏表示的二值图像超分辨率重建算法[J].计算机技术与发展,2017,27(12):43.[doi:10.3969/ j. issn.1673-629X.2017.12.010]
　SHEN Yi-ling,ZHAO Ming-zhe,LI Qiang-yi,et al.A Super-resolution Reconstruction Algorithm for Binary Image Based on Sparse Representation[J].,2017,27(04):43.[doi:10.3969/ j. issn.1673-629X.2017.12.010]
[17]周 航,董西伟,荆晓远.基于优化字典设计的 MOD 字典学习算法[J].计算机技术与发展,2018,28(01):56.[doi:10.3969/ j. issn.1673-629X.2018.01.012]
　ZHOU Hang,DONG Xi-wei,JING Xiao-yuan.Design of MOD Dictionary Learning Algorithm Based onOptimized Dictionary[J].,2018,28(04):56.[doi:10.3969/ j. issn.1673-629X.2018.01.012]
[18]王 君,唐贵进,刘小花,等.基于 Criminisi 的结构组稀疏表示图像修复算法[J].计算机技术与发展,2020,30(03):24.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 005]
　WANG Jun,TANG Gui-jin,LIU Xiao-hua,et al.Criminisi-based Structural Group Sparse Representation for Image Inpainting[J].,2020,30(04):24.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 005]
[19]张敬尊,张睿哲,徐光美,等.稀疏表示模型及高光谱遥感应用研究[J].计算机技术与发展,2020,30(10):173.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 031]
　ZHANG Jing-zun,ZHANG Rui-zhe,XU Guang-mei,et al.Research of Sparse Models and Applications in Hyperspectral Images[J].,2020,30(04):173.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 031]

更新日期/Last Update: 2019-04-10