PSW statistical LSB image steganalysis

Shojae Chaeikar, Saman and Zamani, Mazdak and Abdul Manaf, Azizah and Zeki, Akram M. (2017) PSW statistical LSB image steganalysis. Multimedia Tools and Applications, 77 (1). pp. 805-835. ISSN 1380-7501 E-ISSN 1573-7721

	PDF - Published Version Restricted to Repository staff only Download (2MB) | Request a copy
Preview	PDF - Supplemental Material Download (75kB) | Preview

Abstract

Steganography is the art and science of producing covert communications by concealing secret messages in apparently innocent media, while steganalysis is the art and science of detecting the existence of these. This manuscript proposes a novel blind statistical steganalysis technique to detect Least Significant Bit (LSB) flipping image steganography. It shows that the technique has a number of major advantages. First, a novel method of pixel color correlativity analysis in Pixel Similarity Weight (PSW). Second, filtering out image pixels according to their statistically detected suspiciousness, thereby excluding neutral pixels from the steganalysis process. Third, ranking suspicious pixels according to their statistically detected suspiciousness and determining the influence of such pixels based on the level of detected anomalies. Fourth, the capability to classify and analyze pixels in three pixel classes of flat, smooth and edgy, thereby enhancing the sensitivity of the steganalysis. Fifth, achieving an extremely high efficiency level of 98.049% in detecting 0.25bpp stego images with only a single dimension analysis.

Item Type:	Article (Journal)
Additional Information:	6153/62992
Uncontrolled Keywords:	Blind steganalysis . Statistical steganalysis . Pixel similarity. Color correlativity . Flipping steganography. Image steganalysis . LSB . Machine learning . Support vectormachine
Subjects:	T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Information and Communication Technology > Department of Information System Kulliyyah of Information and Communication Technology > Department of Information System
Depositing User:	Akram M Zeki
Date Deposited:	19 Feb 2019 10:12
Last Modified:	12 Jul 2019 16:42
URI:	http://irep.iium.edu.my/id/eprint/62992

Actions (login required)

View Item