Detecting Steganography in images from mobile stego apps using random statistical properties

Date
2021
Authors
Guan, Yong
Chen, Wenhao
Lin, Li
Martin, Abby
Maxion, Roy
Newman, Jennifer
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Copyright 2021, The Authors
Altmetrics
Authors
Guan, Yong
Person
Research Projects
Organizational Units
Journal Issue
Series
Department
MathematicsElectrical and Computer EngineeringCenter for Statistics and Applications in Forensic Evidence
Abstract
Steganography apps on smartphones have seen increasing use over the last several years. Using a stego app, a secret message – payload - can be hidden inside a photo on your phone, and this seemingly-innocent picture - a stego image - passed to a recipient. However, detection of hidden messages inside images, or steganalysis, is more difficult. Since there is no universal detection method yet available, steganalysis must be targeted for images created under specified conditions. In particular, there are no software tools in existence that detect stego images produced from any mobile stego apps. A few academic algorithms have shown that it is indeed possible to detect such stego images, using image data available in the StegoAppDB database, but no software tool has been developed. this project presents proof-of-concept for detecting stego images created from mobile apps, for a class of embedding algorithms. Specifically, our goal is to detect unencrypted payload hidden inside an image by mobile stego apps, applying tests of randomness to the bit sequences in the image’s lower bit planes.
Comments
The following was presented at International Conference on Forensic and Investigative Science (ICFFIS) (2021). Posted with permission of CSAFE.
Description
Keywords
Citation
DOI
Source
Collections