Reversible data hiding in encrypted Images

Other Title(s)

حماية حق الملكية للصور المشفرة باستخدام خوارزمية الدمغ الرقمي

Dissertant

al-Kittawi, Nur Husayn

Thesis advisor

al-Haj, Ali

Comitee Members

al-Qarallah, Isam A.
al-Tawayha, Ahmad
Jafar, Iyad

University

Princess Sumaya University for Technology

Faculty

King Abdullah II Faculty of Engineering

Department

Department of Electrical Engineering

University Country

Jordan

Degree

Master

Degree Date

2016

English Abstract

Nowadays, a large number of digital images are being transmitted for different applications using public networks.

In those networks, copying, distributing, and modifying images contents become very easy.

The need to achieve confidentiality, integrity, and authenticity for transmitted images has become an urgent requirement.

Indeed, transmitted images must be secured so that only the authorized receivers view or modify their contents.

The encryption techniques convert digital images into unintelligible data for privacy protection.

Data-hiding techniques embed additional information into digital images for copyright protection, content annotation or covert communication.

Recently, techniques that combine the encryption and data-hiding have attracted considerable research interest, because the data hider is not always the same as the data owner.

Hence data hiding schemes for encrypted images need to be developed.

Data hiding schemes in encrypted image should be reversible and cannot introduce any permanent distortion into the image because this may lead to decryption errors.

This thesis proposes two novel algorithms to reversibly hide data into encrypted grayscale image in a separable manner.

Separable means, if the receiver has the data hiding key only, then the hidden data can be extracted without decrypting the image.

On the other hand, if the receiver has the encryption key only, it is possible to decrypt the image, but not to extract the hidden data.

If the receiver has both keys, then the hidden data can be extracted and the original plain image can be restored.

In the first algorithm, the image owner encrypts the image then a data hider hides two watermarks in the encrypted image.

One watermark is embedded by flipping the two middle bits of encrypted image pixels, which are selected based on data hiding key.

The other watermark is embedded by histogram shifting method applied on the watermarked image.

At the receiver side, the second watermark is extracted first using the inverse histogram method to verify the image integrity and authenticity.

Then, while the encrypted image containing the first watermark; a content user with the decryption key may decrypt the image to obtain the principal original content.

If someone receives the decrypted image and has the data-hiding key, he can also successfully extract the data and perfectly recover the original image by exploiting the spatial correlation in the original image.

The second algorithm also hides two watermarks in a given encrypted image.

The first watermark is embedded by replacing selected encrypted image pixels based on data hiding key, and the second watermark is embedded in the watermarked encrypted image using the RDH histogram shifting method.

At the receiving end, if the receiver has the data-hiding key, he can extract the embedded data; however, the image content will remain encrypted.

On the other hand, if the receiver has the encryption key, he can decrypt the received image to obtain an image similar to the original one, but would not be able to extract the additional data.

Finally, if the receiver has both the data-hiding key and the encryption key, then he can extract the additional data and recover the original content.

The experimental results demonstrate that the two proposed algorithms have high embedding capacity, high visual image quality, and high entropy.

The maximum achieved capacities are 0.0648 for algorithm I, and 0.0782 for algorithm II.

The high visual image quality achieved by the two algorithms is reflected by the 40 dB PSNR value achieved by algorithm I, and the 44 dB PSNR value achieved by algorithm II.

The entropy provided by the two algorithms is high and approaches 7.8 bits/pixel for both algorithms.

Finally, the two algorithms are competitive with other recent work in the field.

As a matter of fact, the embedding capacity of the encrypted image is much higher than the existing reversible data hiding techniques in encrypted image since two watermarks are embedded in the encrypted image by each algorithm.

Main Subjects

Electronic engineering

No. of Pages

58

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-720847