JPEG steganalysis using HBCL statistics and FR index

Downloads

Downloads per month over past year

Bhat, V.H. and Krishna, S. and Shenoy, P.D. and Venugopal, K.R. and Patnaik, L.M. (2010) JPEG steganalysis using HBCL statistics and FR index. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 6122 L. pp. 105-112. ISSN 03029743

[img] Text
JPEG Steganalysis Using HBCL Statistics and FR Index.pdf
Restricted to Registered users only

Download (348kB)
Official URL: https://doi.org/10.1007/978-3-642-13601-6_13

Abstract

This paper introduces a new statistical model for blind steganalysis of JPEG images. The functionals used to build this model are based on Huffman Bit Code Lengths (HBCL statistics) and the file size to image resolution ratio (FR Index). JPEG images spanning a wide range of resolutions were used to create a 'stego-image' database employing three embedding schemes - the advanced Least Significant Bit encoding technique, JPEG Hide-and-Seek and Model Based Steganography. Existing blind steganalysis techniques mostly involve the analyses of generalized category attacks and the higher order statistics. This work builds an effective neural network prediction model using HBCL statistics and FR Index, which are not yet explored by steganalysts. The experimental results proved to be efficient over a diverse image database and several payloads. © 2010 Springer-Verlag.

Item Type: Article
Additional Information: cited By 2; Conference of Pacific Asia Workshop on Intelligence and Security Informatics, PAISI 2010 ; Conference Date: 21 June 2010 Through 21 June 2010; Conference Code:81100
Uncontrolled Keywords: Bit codes; Blind steganalysis; Encoding techniques; File sizes; Functionals; Higher order statistics; Huffman; Image database; JPEG image; Least significant bits; Model-based; Neural network prediction model; Statistical models; Statistical Steganalysis; Steganalysis; Stego image, Cryptography; Image coding; Image resolution; Information science; Mathematical models; Statistics; Steganography, Neural networks
Subjects: Faculty of Engineering > Computer Science & Information Science Engineering
Divisions: University Visvesvarayya College of Engineering > Department of Computer Science and Information Science Engineering
Depositing User: Mr. Kirana Kumar D
Date Deposited: 22 Mar 2016 09:18
Last Modified: 22 Mar 2016 09:18
URI: http://eprints-bangaloreuniversity.in/id/eprint/2177

Actions (login required)

View Item View Item