Text hiding in RGBA images using the alpha channel and the indicator method

Other Title(s)

إخفاء النصوص في صور من نوع RGBA باستخدام قناة ألفا و طريقة المؤشر

Dissertant

al-Sarayirah, Ghayth Salim

Thesis advisor

al-Jarrah, Muzaffar

Comitee Members

Ahmad, Ahmad K.
al-Sadi, Jihad

University

Middle East University

Faculty

Faculty of Information Technology

Department

Computer Science Department

University Country

Jordan

Degree

Master

Degree Date

2014

English Abstract

Steganography is the art and science of protecting confidential information through concealing its existence.

It is an ancient art of hiding secret messages in forms or media that cannot be observed by an adversary.

The field of Steganography has been revived lately to become an important tool for secure transmission of secret messages.

This thesis investigates the hiding of text messages and documents within the alpha channel of RGBA color images.

The proposed model consists of two phases.

In the first phase the secret text is stored as bits in the LSB part of the alpha channel.

The number of bits stored in each alpha channel of a pixel is variable, in order to make it harder for an attacker to detect.

The RGB color channels of a pixel are used as a guide or indicator to select the number of bits to store in the alpha channel of the same pixel.

The numbers of hidden bits per pixels are 1, 2, or 3, depending on the indicator value, which gives an average of 3 bpp.

This phase is implemented in two algorithms: Embed, to hide a secret text document inside a cover image of RBGA type, and Extract, to recover the hidden text document from the stego image produced during embedding.

The purpose of the second phase is to separate the alpha channel from the RGB channels of the stego image, and to attach the alpha channel to a different image, a semi-stego, with different RGB channels values, in other words to Swap the alpha channel of the two un-related images.

The resulting semi-stego RGBA image will show no relationship between the color channels and the alpha channel.

The semi-stego image is transmitted to the destination.

At the destination, the semi-stego goes through another swap, in which the alpha channel is separated and attached to a copy of the original cover image which the destination will have for this purpose.

The final part of this phase is the extraction of the secret message from the re-created stego RGBA image.

The swapping operation is implemented in the Swap algorithm.

The experimental work tested the hiding and extraction of text documents of various sizes, up to the maximum hiding capacity.

The stego images showed no visual quality degradation even with using the maximum hiding capacity.

A comparison is made of the PSNR metric between the proposed model and that of a model that uses hiding in the RGB channels, using the same color images, and an equal size secret messages.

The results showed a similar or improved PSNR value for most of the test pairs of image / text.

The contributions of this work is three fold: (i) hiding in the alpha channel, leaving the color channels un-altered, which gives better un-detectability compared to just RGB, (ii) Using the color channel as an indicator to store in the alpha channel, and (iii) Using the Swap operation to exchange the alpha channel between the original stego image and the semi-stego image.

The proposed algorithms have been implemented in Matlab 2012b.

A possible application of the proposed model is in the hiding of confidential documents transmitted over public networks.

The thesis ends with conclusions and suggestions for further work.

Main Subjects

Information Technology and Computer Science

No. of Pages

60

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-699705