A feature selection methodology for steganalysis

Authors:
Yoan Miche;Benoit Roue;Amaury Lendasse;Patrick Bas
Affiliations:
Laboratory of Computer and Information Science, Helsinki University of Technology, Hut, Finland;Laboratoire des Images et des Signaux de Grenoble, Saint Martin d'Hères, France;Laboratory of Computer and Information Science, Helsinki University of Technology, Hut, Finland;Laboratory of Computer and Information Science, Helsinki University of Technology, Hut, Finland
Venue:
MRCS'06 Proceedings of the 2006 international conference on Multimedia Content Representation, Classification and Security
Year:
2006

Citing 5
Cited 6

Detecting Hidden Messages Using Higher-Order Statistics and Support Vector Machines

IH '02 Revised Papers from the 5th International Workshop on Information Hiding
Improving LSB steganalysis using marginal and joint probabilistic distributions

Proceedings of the 2004 workshop on Multimedia and security
Defending against statistical steganalysis

SSYM'01 Proceedings of the 10th conference on USENIX Security Symposium - Volume 10
Towards multi-class blind steganalyzer for JPEG images

IWDW'05 Proceedings of the 4th international conference on Digital Watermarking
Detection of LSB steganography via sample pair analysis

IEEE Transactions on Signal Processing

Least significant bit steganography detection with machine learning techniques

Proceedings of the 2007 international workshop on Domain driven data mining
On completeness of feature spaces in blind steganalysis

Proceedings of the 10th ACM workshop on Multimedia and security
Impact of feature selection in classification for hidden channel detection on the example of audio data hiding

Proceedings of the 10th ACM workshop on Multimedia and security
Benchmarking for Steganography

Information Hiding
Advantages of using feature selection techniques on steganalysis schemes

IWANN'07 Proceedings of the 9th international work conference on Artificial neural networks
Audio steganalysis with Hausdorff distance higher order statistics using a rule based decision tree paradigm

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a methodology to select features before training a classifier based on Support Vector Machines (SVM). In this study 23 features presented in [1] are analysed. A feature ranking is performed using a fast classifier called K-Nearest-Neighbours combined with a forward selection. The result of the feature selection is afterward tested on SVM to select the optimal number of features. This method is tested with the Outguess steganographic software and 14 features are selected while keeping the same classification performances. Results confirm that the selected features are efficient for a wide variety of embedding rates. The same methodology is also applied for Steghide and F5 to see if feature selection is possible on these schemes.