Damageless information hiding technique using neural network

Kensuke Naoe, Yoshiyasu Takefuji

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a new information hiding technique without embedding any information into the target content by using neural network trained on frequency domain. Proposed method can detect a hidden bit codes from the content by processing the selected feature subblocks into the trained neural network. Hidden codes are retrieved from the neural network only with the proper extraction key provided. The extraction key, in proposed method, are the coordinates of the selected feature subblocks and the network weights generated by supervised learning of neural network. The supervised learning uses the coefficients of the selected feature subblocks as set of input values and the hidden bit patterns are used as teacher signal values of neural network. With our proposed method, we are able to introduce a information hiding scheme with no damage to the target content.

Original languageEnglish
Title of host publicationProceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008
Pages253-256
Number of pages4
DOIs
Publication statusPublished - 2008
Event2008 4th International Conference on Intelligent Information Hiding and Multiedia Signal Processing, IIH-MSP 2008 - Harbin, China
Duration: 2008 Aug 152008 Aug 17

Other

Other2008 4th International Conference on Intelligent Information Hiding and Multiedia Signal Processing, IIH-MSP 2008
CountryChina
CityHarbin
Period08/8/1508/8/17

Fingerprint

Neural networks
Supervised learning
Processing

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Graphics and Computer-Aided Design
  • Signal Processing

Cite this

Naoe, K., & Takefuji, Y. (2008). Damageless information hiding technique using neural network. In Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008 (pp. 253-256). [4604051] https://doi.org/10.1109/IIH-MSP.2008.220

Damageless information hiding technique using neural network. / Naoe, Kensuke; Takefuji, Yoshiyasu.

Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008. 2008. p. 253-256 4604051.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Naoe, K & Takefuji, Y 2008, Damageless information hiding technique using neural network. in Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008., 4604051, pp. 253-256, 2008 4th International Conference on Intelligent Information Hiding and Multiedia Signal Processing, IIH-MSP 2008, Harbin, China, 08/8/15. https://doi.org/10.1109/IIH-MSP.2008.220
Naoe K, Takefuji Y. Damageless information hiding technique using neural network. In Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008. 2008. p. 253-256. 4604051 https://doi.org/10.1109/IIH-MSP.2008.220
Naoe, Kensuke ; Takefuji, Yoshiyasu. / Damageless information hiding technique using neural network. Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008. 2008. pp. 253-256
@inproceedings{c3e237c8fdf94a50b5afe8dad2fddd89,
title = "Damageless information hiding technique using neural network",
abstract = "In this paper, we propose a new information hiding technique without embedding any information into the target content by using neural network trained on frequency domain. Proposed method can detect a hidden bit codes from the content by processing the selected feature subblocks into the trained neural network. Hidden codes are retrieved from the neural network only with the proper extraction key provided. The extraction key, in proposed method, are the coordinates of the selected feature subblocks and the network weights generated by supervised learning of neural network. The supervised learning uses the coefficients of the selected feature subblocks as set of input values and the hidden bit patterns are used as teacher signal values of neural network. With our proposed method, we are able to introduce a information hiding scheme with no damage to the target content.",
author = "Kensuke Naoe and Yoshiyasu Takefuji",
year = "2008",
doi = "10.1109/IIH-MSP.2008.220",
language = "English",
isbn = "9780769532783",
pages = "253--256",
booktitle = "Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008",

}

TY - GEN

T1 - Damageless information hiding technique using neural network

AU - Naoe, Kensuke

AU - Takefuji, Yoshiyasu

PY - 2008

Y1 - 2008

N2 - In this paper, we propose a new information hiding technique without embedding any information into the target content by using neural network trained on frequency domain. Proposed method can detect a hidden bit codes from the content by processing the selected feature subblocks into the trained neural network. Hidden codes are retrieved from the neural network only with the proper extraction key provided. The extraction key, in proposed method, are the coordinates of the selected feature subblocks and the network weights generated by supervised learning of neural network. The supervised learning uses the coefficients of the selected feature subblocks as set of input values and the hidden bit patterns are used as teacher signal values of neural network. With our proposed method, we are able to introduce a information hiding scheme with no damage to the target content.

AB - In this paper, we propose a new information hiding technique without embedding any information into the target content by using neural network trained on frequency domain. Proposed method can detect a hidden bit codes from the content by processing the selected feature subblocks into the trained neural network. Hidden codes are retrieved from the neural network only with the proper extraction key provided. The extraction key, in proposed method, are the coordinates of the selected feature subblocks and the network weights generated by supervised learning of neural network. The supervised learning uses the coefficients of the selected feature subblocks as set of input values and the hidden bit patterns are used as teacher signal values of neural network. With our proposed method, we are able to introduce a information hiding scheme with no damage to the target content.

UR - http://www.scopus.com/inward/record.url?scp=54049132560&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=54049132560&partnerID=8YFLogxK

U2 - 10.1109/IIH-MSP.2008.220

DO - 10.1109/IIH-MSP.2008.220

M3 - Conference contribution

AN - SCOPUS:54049132560

SN - 9780769532783

SP - 253

EP - 256

BT - Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008

ER -