APSIPA Transactions on Signal and Information Processing > Vol 12 > Issue 1

Reversible Data Hiding in Compressible Encrypted Images with Capacity Enhancement

Ryota Motomura, Chiba University, Japan, Shoko Imaizumi, Chiba University, Japan, imaizumi@chiba-u.jp , Hitoshi Kiya, Tokyo Metropolitan University, Japan
 
Suggested Citation
Ryota Motomura, Shoko Imaizumi and Hitoshi Kiya (2023), "Reversible Data Hiding in Compressible Encrypted Images with Capacity Enhancement", APSIPA Transactions on Signal and Information Processing: Vol. 12: No. 1, e31. http://dx.doi.org/10.1561/116.00000014

Publication Date: 04 Jul 2023
© 2023 R. Motomura, S. Imaizumi and H. Kiya
 
Subjects
 
Keywords
Reversible data hiding in encrypted imageshiding capacitycompressible image encryptionencryption-then-compression systemlossless compression
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 1008 times

In this article:
Introduction 
Related Work 
Proposed Method 
Experimental Results 
Conclusion 
References 

Abstract

In this paper, we propose a method for high-capacity reversible data hiding in compressible encrypted images. While reversible data hiding in encrypted images (RDH-EI) has been actively studied, most previous research in this field has focused on the hiding capacity. We previously proposed an RDH-EI method that achieves both a high hiding capacity and high compression performance simultaneously. The method can effectively compress marked encrypted images using lossless image coding standards with a hiding capacity of up to about 1~bpp. In addition, there is an option to decrypt marked encrypted images without data extraction, so marked images can be obtained, still containing the embedded data, i.e., payload. Without losing the advantages of the previous method, we propose an extended method, in which the highest frequency bin is constantly used for data hiding. Consequently, the hiding capacity is further enhanced up to 2.45 bpp for each color component in the proposed method. Through our experiments, we prove the superiority of the proposed method.

DOI:10.1561/116.00000014