Automatic Recovery of Hidden Image from Image Steganography Using DNN and Local Entropy Features

J. H. Lee, D. Y. Kang, J. E. Lee, S. H. Lee, J. I. Park

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Image steganography hides secret information in an image called cover image so naturally that the other users can not recognize the existence of information in the revealed image. This paper deals with an approach to recover the hidden image information from image steganography. The proposed approach investigates that the decoded hidden image information is a normal image or not. The normal and incorrectly decoded abnormal images have been trained using a deep neural network model and entropy features. The discrimination is processed with image patches since the information may be partially embedded in the cover image. The experiments are performed with respect to the various data capacities. The proposed approach discriminates and recovers the hidden image information automatically from a tremendously large number of steganography encoding methods.

Original languageEnglish
Title of host publicationITC-CSCC 2020 - 35th International Technical Conference on Circuits/Systems, Computers and Communications
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages440-445
Number of pages6
ISBN (Electronic)9784885523281
StatePublished - 2020 Jul
Event35th International Technical Conference on Circuits/Systems, Computers and Communications, ITC-CSCC 2020 - Nagoya, Japan
Duration: 2020 Jul 32020 Jul 6

Publication series

NameITC-CSCC 2020 - 35th International Technical Conference on Circuits/Systems, Computers and Communications

Conference

Conference35th International Technical Conference on Circuits/Systems, Computers and Communications, ITC-CSCC 2020
Country/TerritoryJapan
CityNagoya
Period20/07/320/07/6

Keywords

  • Data hiding
  • Deep neural network
  • Image entropy
  • Image steganography
  • Steganalysis

Fingerprint

Dive into the research topics of 'Automatic Recovery of Hidden Image from Image Steganography Using DNN and Local Entropy Features'. Together they form a unique fingerprint.

Cite this