|本期目录/Table of Contents|

[1]赵 娜,赵彤洲*,邹 冲,等.稀疏表示中字典学习的影响因子研究[J].武汉工程大学学报,2017,39(03):267-272.[doi:10. 3969/j. issn. 1674?2869. 2017. 03. 011]
 HAO Na,ZHAO Tongzhou*,ZOU Chong,et al.nfluence Factors of Dictionary Learning in Sparse Representation[J].Journal of Wuhan Institute of Technology,2017,39(03):267-272.[doi:10. 3969/j. issn. 1674?2869. 2017. 03. 011]
点击复制

稀疏表示中字典学习的影响因子研究(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
39
期数:
2017年03期
页码:
267-272
栏目:
机电与信息工程
出版日期:
2017-06-24

文章信息/Info

Title:
nfluence Factors of Dictionary Learning in Sparse Representation
文章编号:
20170311
作者:
赵 娜12赵彤洲12*邹 冲12刘 莹12蔡敦波12
1. 武汉工程大学计算机科学与工程学院,湖北 武汉 430205;2. 智能机器人湖北省重点实验室,湖北 武汉 430205
Author(s):
HAO Na12ZHAO Tongzhou12*ZOU Chong12LIU Ying12CAI Dunbo12
1. School of Computer Science and Engineering,Wuhan Institute of Technology,Wuhan 430205, China;2. Hubei Key Laboratory of Intelligent Robot(Wuhan Institute of Technology),Wuhan 430205, China
关键词:
稀疏表示字典学习字典精度图像质量评价指标
Keywords:
sparse representation dictionary learning dictionary accuracy image quality assessment index
分类号:
TP391
DOI:
10. 3969/j. issn. 1674?2869. 2017. 03. 011
文献标志码:
A
摘要:
研究了稀疏表示中影响字典矩阵构建质量的关键因素,并实现了关键因子定量化表示. 分别对图像数量、取块大小、字典列数和取块步长等因子进行参数调整并生成字典矩阵,结合系数矩阵对原始图像重构,以峰值信噪比和结构相似性索引测量这两种质量评价指标作为字典质量的评估依据. 实验以CMU_PIE_Face数据库为数据源,结果表明当图像数量为500张、取块大小为4个像素点、字典列数为512维、取块步长为2个像素点时,所得到的字典具备对原始图像的最佳表示能力. 因此,稀疏表示中关键因子的定量化表示可加速字典学习过程且简化模型复杂度,提高字典抽象层质量,具备更强的图像表现力.
Abstract:
We studied the key factors influencing the construction quality of dictionary matrix in sparse representation, and represented them quantitatively. The factors such as the number of images, patch size, dictionary columns and patch step were adjusted as parameters and the dictionary matrix was generated. Combined with the coefficient matrix, the original image was reconstructed, and the dictionary quality was evaluated by using the image quality assessment indices of peak signal to noise ratio and structural similarity index metric. Experiments on CMU_PIE_Face database demonstrate that the resulting dictionary has the best ability to represent the original image at image numbers of 500, patch size of 4 px, dictionary columns of 512 and patch step of 2 px. We found that the quantitative representation of key factors in sparse representation can accelerate the dictionary learning process, simplify the complexity of the model, improve the quality of the dictionary abstraction layer, and show stronger image expression.

参考文献/References:

[1] OLSHAUSEN B A, FIELD D J. Emergence of simple-cell receptive field properties by learning a sparse code for natural images [J]. Nature, 1996, 381(6583): 607-609. [2] LI S, YIN H, FANG L. Group-sparse representation with dictionary learning for medical image denoising and fusion [J]. IEEE Transactions on Biomedical Engineering, 2012, 59(12): 3450-3459. [3] DONG W, ZHANG L, SHI G, et al. Nonlocally centralized sparse representation for image restoration [J]. IEEE Transactions on Image Processing, 2013, 22(4): 1620-1630. [4] WAGNER A, WRIGHT J, GANESH A, et al. Toward a practical face recognition system: robust alignment and illumination by sparse representation [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2012, 34(2): 372-386. [5] 朱杰, 杨万扣, 唐振民. 基于字典学习的核稀疏表示人脸识别方法[J]. 模式识别与人工智能, 2012, 25(5):859-864. ZHU J, YANG W K, TANG Z M. A dictionary learning based kernel sparse representation method for face recognition [J]. Pattern Recognition and Artificial Intelligence, 2012, 25(5):859-864. [6] SHEKHAR S, PATEL V M, NASRABADI N M, et al. Joint sparse representation for robust multimodal biometrics recognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(1): 113-126. [7] 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. [8] MAIRAL J, BACH F, PONCE J, et al. Online learning for matrix factorization and sparse coding [J]. Journal of Machine Learning Research, 2010, 11(1): 19-60. [9] 练秋生,石保顺, 陈书贞.字典学习模型、算法及其应用研究进展[J]. 自动化学报,2015, 41(2):240-260. LIAN Q S, SHI B S, CHEN S Z. Research advances on dictionary learning models, algorithms and applications [J]. Acta Automatica Sinica, 2015, 41(2):240-260. [10] 韦仙, 康睿丹. 基于降维压缩法的图像重构[J]. 武汉工程大学学报, 2015, 37(12):69-74. WEI X, KANG R D. Image reconstruction based on dimension reduction and compression technology [J]. Journal of Wuhan Institute of Technology, 2015, 37(12):69-74. [11] CANDES E J,TAO T. Decoding by linear programming [J]. IEEE Transactions on Information Theory, 2005, 51(12): 4203-4215. [12] 欧卫华. 基于稀疏表示和非负矩阵分解的部分遮挡人脸识别研究 [D]. 武汉:华中科技大学, 2014. [13] 李洪均, 谢正光, 胡伟, 等. 字典原子优化的图像稀疏表示及其应用 [J]. 东南大学学报(自然科学版), 2014 (1): 116-122. LI H J, XIE Z G, HU W,et al. Optimization of dictionary atoms in image sparse representations and its application[J]. Journal of Southeast University (Natural Science), 2014 (1) : 116-122. [14] 吴双, 邱天爽, 高珊. 基于在线字典学习的医学图像特征提取与融合 [J]. 中国生物医学工程学报, 2014, 33(3): 283-288. WU S, QIU T S, GAO S. Medical image features extraction and fusion based on online dictionary learning[J]. Chinese Journal of Biomedical Engineering, 2014, 33(3) : 283-288 . [15] 霍雷刚. 图像处理中的块先验理论及应用研究 [D]. 西安:西安电子科技大学, 2015.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2016-12-06基金项目:国家自然科学基金(61103136);武汉工程大学创新基金(CX2015057);武汉工程大学创新基金(CX2016070)作者简介:赵 娜,硕士研究生. E-mail:zhaona_wit@163.com*通讯作者:赵彤洲,硕士,副教授. E-mail:zhao_tongzhou@126.com
更新日期/Last Update: 2017-06-23