Mixed Norm Regularized Discrimination for Image Steganalysis

被引:3
作者
Chen, Guoming [1 ]
Chen, Qiang [1 ]
Zhang, Dong [2 ]
机构
[1] Guangdong Univ Educ, Dept Comp Sci, Guangzhou 510310, Guangdong, Peoples R China
[2] Sun Yat Sen Univ, Guangzhou 510275, Guangdong, Peoples R China
来源
SENSING AND IMAGING | 2015年 / 16卷 / 01期
基金
中国国家自然科学基金;
关键词
L-p; L-q norm; Feature selection; Steganalysis;
D O I
10.1007/s11220-015-0120-5
中图分类号
TH7 [仪器、仪表];
学科分类号
0804 ; 080401 ; 081102 ;
摘要
The purpose of image steganalysis is to detect the presence of hidden messages in cover images. Steganalysis can be considered as a pattern recognition process to decide which class a test image belongs to: the innocent photographic image or the stego-image. This paper presents a definition of mixed L-p,L-q matrix norm as an extension of L-2,L-1 matrix norm. We incorporate discriminative mixed L-p,L-q matrix norm analysis to select the features which best preserve the data distribution, e.g., manifold structure, of the whole feature sets. Experiments on different data sets verify the effectiveness of the proposed approach and the selected features are more discriminate.
引用
收藏
页数:14
相关论文
共 12 条
[1]   Local and Global Discriminative Learning for Unsupervised Feature Selection [J].
Du, Liang ;
Shen, Zhiyong ;
Li, Xuan ;
Zhou, Peng ;
Shen, Yi-Dong .
2013 IEEE 13TH INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), 2013, :131-140
[2]   Mixed-norm estimates for the M/EEG inverse problem using accelerated gradient methods [J].
Gramfort, Alexandre ;
Kowalski, Matthieu ;
Haemaelaeinen, Matti .
PHYSICS IN MEDICINE AND BIOLOGY, 2012, 57 (07) :1937-1961
[3]  
He R, 2012, PROC CVPR IEEE, P2504, DOI 10.1109/CVPR.2012.6247966
[4]   Joint Embedding Learning and Sparse Regression: A Framework for Unsupervised Feature Selection [J].
Hou, Chenping ;
Nie, Feiping ;
Li, Xuelong ;
Yi, Dongyun ;
Wu, Yi .
IEEE TRANSACTIONS ON CYBERNETICS, 2014, 44 (06) :793-804
[5]   lp-lq Penalty for Sparse Linear and Sparse Multiple Kernel Multitask Learning [J].
Rakotomamonjy, Alain ;
Flamary, Remi ;
Gasso, Gilles ;
Canu, Stephane .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2011, 22 (08) :1307-1320
[6]   Hessian Semi-Supervised Sparse Feature Selection Based on L2,1/2-Matrix Norm [J].
Shi, Caijuan ;
Ruan, Qiuiqi ;
An, Gaoyun ;
Zhao, Ruizhen .
IEEE TRANSACTIONS ON MULTIMEDIA, 2015, 17 (01) :16-28
[7]  
Shi Y., P IEEE INT C INF TEC, V1, P768
[8]  
Wang L. P., 2013, IEEE COMPUTER VISION
[9]   Optimal Feature Selection for Robust Classification via l2,1-Norms Regularization [J].
Wen, Jiajun ;
Lai, Zhihui ;
Wong, Wai Keung ;
Cui, Jinrong ;
Wan, Minghua .
2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2014, :517-521
[10]   Sparse Representation Preserving for Unsupervised Feature Selection [J].
Yan, Hui ;
Jin, Zhong ;
Yang, Jian .
2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2014, :1574-1578