Welcome to the InfoSci Platform
Information Hiding by Machine Learning: A Method of Key Generation for Information Extracting Using Neural Network

Information Hiding by Machine Learning: A Method of Key Generation for Information Extracting Using Neural Network

Kensuke Naoe, Hideyasu Sasaki, Yoshiyasu Takefuji
Copyright: © 2011 |Volume: 2 |Issue: 1 |Pages: 28
ISSN: 1947-9344|EISSN: 1947-9352|EISBN13: 9781613508794|DOI: 10.4018/ijoci.2011010102
Cite Article Cite Article

MLA

Naoe, Kensuke, et al. "Information Hiding by Machine Learning: A Method of Key Generation for Information Extracting Using Neural Network." IJOCI vol.2, no.1 2011: pp.21-48. http://doi.org/10.4018/ijoci.2011010102

APA

Naoe, K., Sasaki, H., & Takefuji, Y. (2011). Information Hiding by Machine Learning: A Method of Key Generation for Information Extracting Using Neural Network. International Journal of Organizational and Collective Intelligence (IJOCI), 2(1), 21-48. http://doi.org/10.4018/ijoci.2011010102

Chicago

Naoe, Kensuke, Hideyasu Sasaki, and Yoshiyasu Takefuji. "Information Hiding by Machine Learning: A Method of Key Generation for Information Extracting Using Neural Network," International Journal of Organizational and Collective Intelligence (IJOCI) 2, no.1: 21-48. http://doi.org/10.4018/ijoci.2011010102

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

In this paper, the authors propose information hiding by machine learning: a method of key generation for information extracting using neural network. The method consists of three layers for information hiding. First, the proposed method prepares feature extraction keys, which are saved by feature extraction attributes like feature coordinates and the region of frequency coefficients. Second, the proposed method prepares hidden patterns in advance to the embedding procedure as a watermark signal of the target contents. Finally, the proposed method generates information extraction keys by using machine learning to output presented hidden patterns. The proper hidden patterns are generated with the proper information extraction key and feature extraction key. In the experiments, the authors show that the proposed method is robust to high pass filtering and JPEG compression. The proposed method contributes to secure visual information hiding without damaging any detailed data of the target content.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.