Elsevier

Information Sciences

Volume 324, 10 December 2015, Pages 197-207
Information Sciences

Image encryption: Generating visually meaningful encrypted images

https://doi.org/10.1016/j.ins.2015.06.049Get rights and content

Abstract

To protect image contents, most existing encryption algorithms are designed to transform an original image into a texture-like or noise-like image which is, however, an obvious visual sign indicating the presence of an encrypted image and thus results in a significantly large number of attacks. To address this problem, this paper proposes a new image encryption concept to transform an original image into a visually meaningful encrypted one. As an example of the implementation of this concept, we introduce an image encryption system. Simulation results and security analysis demonstrate excellent encryption performance of the proposed concept and system.

Introduction

In the past ten years, “Cloud” has become a popular keyword in the computer society. To overcome the limitations in the storage and computation capability of a single computer, cloud computing technology provides individuals and organizations a sufficiently large online space to store multimedia data (e.g. documents, videos and images), and offers people a convenient way to access and share data over the network. Due to the fact that these multimedia data may contain private, valued or even classified information, preventing these important information from leakage becomes an important and urgent issue for individuals and organizations [33], [34]. Image encryption is an efficient tool to provide security to these multimedia data.

Many image encryption algorithms have been proposed to protect images by changing their pixel values and/or locations using different technologies [5], [19], [35]. They can be classified into the frequency-domain and spatial-domain image encryption algorithms. Using security key coefficients, the frequency-domain image encryption algorithms are designed to change image data in the frequency domain or alter the transform function, such as the discrete fractional Fourier transform [18], [21], [26], quantum Fourier transform [31] and reciprocal-orthogonal parametric transform [6]. The spatial-domain image encryption algorithms are based on the famous Substitution-Permutation Network (SPN) that utilizes a substitution process to change image pixel values and a permutation process to change image pixel positions. These permutation and substitution processes are based on different technologies including the advanced encryption standard (AES) [24], P-Fibonacci transform [40], wave transmission [20], elliptic curve ElGamal [8], gray code [36], random grids [10], Latin squares [30] and chaotic systems [9], [17], [38], [41]. Both the spatial-domain and frequency-domain image encryption algorithms are able to protect images with a high level of security [27], [39]. Their output encrypted images are all visually texture-like or noise-like, such as images in Fig. 1(a) and (b). However, since the formats of these encrypted images are limited to noise-like and texture-like, it is easy to distinguish them from normal visually meaningful images. From the security point of view, this texture-like or noise-like feature is an obvious visual sign indicating the presence of an encrypted image [37] that may contain important information. As a result, an encrypted image with similar features as noise-like or texture-like definitely brings increasing people’s attentions and thus leads to a large number of attacks and analysis. These include different types of cryptanalysis, illegal edition, modification or even deletion of image contents. All these increase the possibility of information leakage, loss or modification.

To address these problems, in this paper, we propose a new concept of image encryption to transform original images into visually meaningful encrypted images, such as images in Fig. 1(c). This is because people generally consider these images as normal images rather than encrypted ones. Based on this concept, we introduce an image encryption system. It not only protects images in a normal way that most existing encryption methods do, but also provides an additional visual protection. Simulation results and security analysis are provided.

Section snippets

New concept of image encryption

Most existing image encryption algorithms protect original images by transforming them into texture-like or noise-like encrypted images with a nearly uniform distribution of pixel values. As a result, the encrypted images can withstand different types of attacks, protecting original image information with a high level of security. However, this texture-like or noise-like feature is an obvious visual sign of encrypted images. It definitely catches more people’s attentions and thus brings a

New image encryption system

The key issue of the proposed concept is how to generate VMEIs. Motivated by technologies of image hiding and watermarking, this section introduces a new image encryption system (NIES) as an implementation example of the proposed concept.

Simulation results and performance analysis

This section provides several simulation examples and performance analysis to show the NIES’s encryption performance. In this paper, all reference images are four times larger than the original images to be encrypted.

Security analysis

Generally speaking, the security of an encryption algorithm mainly depends on its security key design [23]. The proposed NIES has a sufficiently large key space and high key sensitivity.

Conclusions

To address the security weakness of most existing encryption algorithms whose texture-like or noise-like encrypted images may bring a large number of attacks and analysis, this paper has introduced a new concept of image encryption to generate visually meaningful encrypted images that usually are considered as normal images rather than encrypted ones. With a large amount of formats of encrypted images, the proposed concept ensures the attackers’ difficulty of correctly distinguishing and

Acknowledgment

This work was supported in part by the Macau Science and Technology Development Fund under grant FDCT/017/2012/A1 and by the Research Committee at University of Macau under grants MYRG2014-00003-FST, MRG017/ZYC/2014/FST, MYRG113(Y1-L3)-FST12-ZYC and MRG001/ZYC/2013/FST.

References (41)

Cited by (195)

View all citing articles on Scopus
View full text