Building a dataset for image steganography

Authors

Chris Woolley, School of Science, Edith Cowan UniversityFollow
Ahmed Ibrahim, School of Science, Edith Cowan UniversityFollow
Peter Hannay, Security Research Institute, Edith Cowan UniversityFollow

Document Type

Conference Proceeding

Publisher

Security Research Institute, Edith Cowan University

Place of Publication

Joondalup, Western Australia

Editor(s)

Professor Craig Valli

ISBN

978-0-6481270-9-3

Abstract

Image steganography and steganalysis techniques discussed in the literature rely on using a dataset(s)created based on cover images obtained from the public domain, through the acquisition of images from Internet sources, or manually. This issue often leads to challenges in validating, benchmarking, and reproducing reported techniques in a consistent manner. It is our view that the steganography/steganalysis research community would benefit from the availability of common datasets, thus promoting transparency and academic integrity. In this research, we have considered four aspects: image acquisition, pre-processing, steganographic techniques, and embedding rate in building a dataset for image steganography.

Comments

Woolley,, C., Ibrahim, A., & Hannay, P. (2017). Building a dataset for image steganography. Paper presented in Valli, C. (Ed.). The Proceedings of 15th Australian Digital Forensics Conference 5-6 December 2017, Edith Cowan University, Perth, Australia.

DOI

10.4225/75/5a83a0541d284