Separable reversible data hiding in encrypted images with improved security and capacity
- PDF / 4,302,628 Bytes
- 20 Pages / 439.642 x 666.49 pts Page_size
- 97 Downloads / 222 Views
Separable reversible data hiding in encrypted images with improved security and capacity Qi Li1 · Bin Yan1
· Hui Li1 · Na Chen1
Received: 8 December 2017 / Revised: 18 May 2018 / Accepted: 22 May 2018 © Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract Reversible data hiding (RDH) has to be conducted in the encrypted images when original images are encrypted for privacy protection in some open environments, including the cloud computing. However, the current RDH algorithms in encrypted images with errorfree decryption may lead to leakage of image content and low embedding rate. In this paper, a novel RDH algorithm for image in encryption domain is proposed. To improve security, we propose a combined block permutation and a stream cipher into the encryption step, which considers data hiding in later steps. We further increase the embedding rate by proposing bit replacement in prediction error. This scheme has the advantages of built-in embedding flag, error-free decryption and high embedding rate. It can be applied to a wide variety of scenarios: If the recipient has only the data-hiding key, he can extract the hidden data but cannot restore the image; If the recipient has only the image encryption key, he can read the distorted image but cannot extract the hidden data; If the recipient has both keys, he can extract the hidden data and restore the original image completely. Keywords Personal privacy · Image encryption · Reversible data hiding · Stream encryption · Block permutation
1 Introduction In recent years, cloud storage and cloud computing are widely used. Since the servers and the users are usually separated from each other, so the data from the users must be uploaded to the cloud server for storage or processing. These data may include sensitive and private information such as personal ID or biometric data. Thus, personal privacy issues have
Bin Yan
[email protected] 1
College of Electronics, Communication and Physics, Shandong University of Science and Technology, Qingdao, 266590, People’s Republic of China
Multimed Tools Appl
become a widespread concern. To protect privacy, the data are usually encrypted before uploading to the cloud server. Without knowing the content of the received data, the server may also need to add additional information for labelling and authentication, etc. A typical application scenario in cloud storage is illustrated in Fig. 1. The user data is an image, which is encrypted in user device 1 (such as a desktop computer), and then is transmitted to cloud server 1. The cloud server 1 embeds a tag into the encrypted image to label the source of this file. The resulting file may need to be transferred to another server, the server 2. In a later time, the user may need to download the file from server 2 to another device, i.e., the device 2, such as his/her cell phone. Then, server 2 needs to extract the hidden data and recover the original encrypted image before sending it to the user. After receiving the recovered image, the user device 2 ca
Data Loading...
