Content uploaded by Md. Mizanur Rahman
Author content
All content in this area was uploaded by Md. Mizanur Rahman on Jan 12, 2019
Content may be subject to copyright.
Intern ational Journ al of Scientific & E ngineering Research, Volume 7, Issue 3, March-2016 864
ISSN 2229-5518
IJSER © 2016
http://www.ijser.org
Secure RGB Image Steganography Based on
Triple-A Algorithm and Pixel Intensity
Md. Mizanur Rahman, Pronab Kumar Mondal, Indrani Mandal, Habiba Sultana
Abstract— Steganography is the art and science of writing hidden messages in such a way that no one, apart from the sender and
intended recipient, even perceives the existence of the hidden message. In this paper, the concept of RGB intensity properties is being
merged with randomization based Triple-A algorithm to increase the hidden capacity of the data-bits. Usually, storing variable number of
bits in each channel (R, G or B) of pixel depend on the actual color values of that pixel. The concept- channels containing lower color
values, can store higher number of data bits. This technique can be applied to RGB images where each pixel is represented by three bytes
to indicate the intensity of Red, Green, and Blue of that pixel. This work shows more effective results, especially the capacity of the data-
bits to be hidden with relation to the RGB image pixels.
Index Terms— Steganography, RGB Bitmap image, Randomization, High Capacity Embedding, Computer Security, Histogram.
—————————— ——————————
1 INTRODUCTION
HE word Steganography originally derived from two
Greek words-Steganos which means “covered or secret”,
and Graphein which means “writing or drawing”. In this
case, Steganography literally means covered writing. Basically,
it is a secret transmission of message between two parties. It is
the practices of encoding or embedding secret information in a
manner that the existence of the information becomes invisi-
ble. This mechanism has been exercising for thousands of
years in various forms. In ancient Greece, the common practic-
es consi sted of etching messages in wooden tablets and cover-
ing them with wax, or tattooing a messenger ’s head after shav-
ing hair and then let his hair grow up before sending him to
the receiver where his hair was shaved again to extract the
massage. Other techniques use invisible ink, microdots, con-
verting channels and character arrangement [1, 2, 3, 4].
Steganography is the art and science to covert communica-
tion. It can be achieved by using carriers like image, audio and
video. In image Steganography the information is hidden ex-
clusively in an image called cover media. After inserting the
secret message it is referred to as stego-image or stego-
medium. A stego key is used for hiding or encoding process to
restrict detection or extraction of the embedded data. So image
Steganography process can be described by the under-
mentioned structure:
Cover image + Embedded message + Stego key = Stego image
The stego-image then sends to the receiver over the public
channel. The receiver can extract the message through using
the stego key which is same as used by the sender. The Fig. 1
shows basic Steganography process.
Fig. 1: Steganography process
Pixel is the smallest unit of an image. Digital images are
stored in computer systems as an array of points (pixels)
where each pixel has three color components: Red, Green, and
Blue (RGB). Each pixel is represented with three bytes to indi-
cate the intensity of these three colors (RGB).
There are many applications of image based Steganogra-
phy, such as-confidential communication and secret data stor-
ing [5, 15], access control system for digital content distribu-
tion [6], digital watermarks [7, 14], and modern printers [9].
Moreover, it can also be used to tag notes to online images and
illegitimate purposes.
This paper endeavors to improve Triple-A algorithm [12] by
adding intensity of pixel that was not considered before. The
basic concept is, the lower intensity of the pixel decides the
number of bits to embed in the cover-image. Because the lower
intensity pixel does not distort the visual quality of pixel, and
it can also store higher number of bits, which has been showed
in Fig. 3.
The rest of the paper is organized as follows: Section 2 pre-
sents some Steganography related existing methods. Section 3
describes our improved technique based on Triple-A algo-
rithm and intensity property. The experimental results and
comparison will be in section 4. Finally, conclusions are given
in section 5.
T
————————————————
• Md. Mizanur Rahman is a Lecturer in Department of Computer Science
and Engineering at Ranada Prasad Shaha University, Dhaka, Bangladesh.
E-mail: mizan173@gmail.com
• Pronab Kumar Mondal and Indrani Mandal are Assistant Professor,
and Habiba Sultana is Lecturer in Computer Science and Engineering
Department, Jatiya Kabi Kazi Nazrul Islam University, Dhaka, Bangla-
desh.
Transmission
Cover Image
Secret Msg
Embedding
Process
Extracting Pro-
cess
Secret Msg
Secret Key
Stego Image
IJSER
Intern ational Journ al of Scientific & E ngineering Research, Volume 7, Issue 3, March-2016 865
ISSN 2229-5518
IJSER © 2016
http://www.ijser.org
2 RELATED WORK
Several methods have been proposed for image based Ste-
ganography where the Least Significant Bit (LSB) substitution
is the simplest one [10, 13, 16]. In LSB, the least significant bit
of each pixel for a specific color channel or for all color chan-
nels is replaced with a bit from the secret data. Although LSB
is simpler than other techniques, it has long standing probabil-
ity of detecting the hidden data. But hiding information
through this algorithm has significant risk. Later Pixel Indica-
tor (PI) based stego system proposed by Adnan Gutub [1, 17]
has substantiated the overall concept. This technique uses the
least two significant bits of one of the channels from Red,
Green or Blue as an indicator for existence of data in the other
two channels. The indicator bits are set randomly in the chan-
nel. But it is hard to predict the embedding capacity through
Pixel Indicator methodology. However, another notable tech-
nique is the Stego Color Cycle (SCC) [11].This SCC technique
uses the RGB images to hide the data in different channels.
That is, it keeps cycling the hidden data between the Red,
Green and Blue channels, utilizing one channel at a cycle time.
This technique is more secure than the LSB but still it is suffers
detecting the cycling pattern that will reveal the secret data.
Also it has less capacity than the LSB. Overall Triple-A tech-
nique uses the same principle of LSB, where the secret is hid-
den in the least significant bits of the pixels, with more ran-
domization in selection of the number of bits and the color
channels that are used [12]. This randomization is expected to
increase the security and capacity of the system.
3 PROPOSED METHOD
In this section, at first we review the Triple-A technique in
[12].Triple-A algorithm taking the message (M), the carrier
image (C), and the password based generated key (K) depend-
ing on password (P), as inputs and produces the message (M)
hidden inside the carrier image (C). This algorithm can be di-
vided into two major parts, Encryption and Hiding as it is
shown in Fig. 2.
In part one the message (M) encrypted using AES algorithm
which will produce Enc (M, K). The key K can be generated
from a set of user passwords each with a specific key using
simple XOR.
In part two, the RGB Image is used as a cover media. It uti-
lizes the advantage of the Bmp images, where every pixel is
independent from the rest of the image file. Enc (M, K) is hid-
den according to Triple-A algorithm which needs to have a
pseudo random number generator (PRNG). The assumption
for PRNG is to give two new random numbers per iteration.
The seeds of these PRNGs namely Seed1 (S1) and Seed2 (S2)
are formed as a function of the Key (K). S1 is restricted to gen-
erate numbers in [0-6] while S2 is restricted to the interval [1-
3].
Fig. 2: Flow chart of Triple-A algorithm
Table 1 shows how S1 random number is used to determine
the component of the RGB image which is used in hiding the
encrypted data Enc (M, K). On the other hand, Table 2 shows
how (S2) random number determines the number of the com-
ponent(s) least significant bits that is used to hide the secret
data. On the same way (S2) random number determines the
number of component bits.
Tabl e 1: Seed1 Random Number Usage
1st PRNG
Random num-
ber
Meaning to the algorithm
0
use R.
1
us e G.
2
use B.
3
us e R G.
4
use RB.
5
use GB.
6
use RGB.
Table 2: Seed2 Random Number Usage
Hiding Encryption
No
Yes
-User enter a massage (M) to hide
-User select a carrier image (C) and password (P)
-Calculate the number of data bits in Enc (M, K)(NB)
-Calculate the maximum number of possible bits in the
carrier (C) to use in order to hide messages (MaxNB)
-Generate key (K) based on the password (P)
-Encrypt the massage (M) using the generated key (k)
to get Enc (M, K)
-
Get seed1 (S1) and seed2 (S2) as a function of the key
(K)
-
Get a pseudo random generated Number for R, G or B
selection using (S1)
-
Get a pseudo random generated Number to choose the
number of data bits to hide in the selected R, G or B
pixel using (S2)
NB<= MaxNB?
-Distribute Enc (M, K) message inside the carrier (C)
End
Start
IJSER
Intern ational Journ al of Scientific & E ngineering Research, Volume 7, Issue 3, March-2016 866
ISSN 2229-5518
IJSER © 2016
http://www.ijser.org
2nd PRNG
Random
number
Meaning to the algorithm
1
use 1 bit of the component(s).
2
use 2 bit of the component(s).
3
use 3 bit of the component(s).
Two tables show that the algorithm may add up to a maxi-
mum of ± 7 to the value of the color component(s) in that pixel.
Also, by combining data from the previous tables, we can see
that the minimum number of bits used in each pixel is 1 if we
use only one bit of one chosen components of the RGB image.
The maximum is 9 bits if we used all the three components
with three bits.
We have improved Triple-A algorithm [12] with respect to
pixel intensity, where color intensity (values of R-G-B) is used
to decide the number of bits to store in pixel. Our technique
ensures a minimum capacity and can accommodate to store
large amount of data. Our idea is that, ‘insignificant’ colors,
significantly more bits can be changed per channel of an RGB
image, because change in lower intensity pixel value has less
visual degradation quality effects. For example according to
the Fig. 3, three pixels (R=255, G=255 and B=255) are generat-
ing White color in (a). In (b), the color component is same as
(a), except that R = 55 generate Green color. In (c), we again set
the 4 LSBs of R to zeros, resulting in a color which seems to be
same as (b).
(a)
(b) (c)
Fig. 3: Effect of colors for changes in the ‘Red’ values
Same scenario occurs for Fig. 3 where if we modify the 4
LSBs (0 to 15) of all pixels (RGB). So, if pixel intensity is less
than 16, it can be modified into 0 to 15 ranges, which will not
be degrade the visual quality of that pixel. So we can also em-
bed up to 4 bits into that pixel.
Our conception is that, in comparison with higher value of
channel the lower color value has less effect on the overall col-
or of pixel. Therefore, more bits can be changed in a channel
having ‘low’ value than a channel with a ‘high’ value. When
RGB component has to be chosen we need to select the lower
color-value channel among the three channels to store higher
number of bits. Therefore, the structure demands to insert the
following steps, and Fig. 4 shows the steps in flow chart.
• Calculate the channel, whose color value is lowest
among the channels.
• Decide the maximum number of data bits to store in
its least significant bits.
• Store the decided number of data-bits in that channel.
Fig. 4: Flow chart of our proposed steps
Fig. 5 demonstrates one example of storing data bits in a
channel. Suppose the data channel (R) is chosen where pixel
value is 22 (00010110) and data bits to embed is 0101. After
embedding data bits pixel output value will be 27 (00011011).
Fig. 5: An example of hiding data bits inside a channe
4 RESULT ANALYSIS AND COMPARISON
This section of the paper describes the experimental results of
our proposed method, and also the comparison with the other
works. Carrier images have been used to hide text message.
Fig. 6 shows an original carrier compared to the same carrier
with secret message using our modified algorithm. From the
first moment, the visual change between the original image
and stego-image cannot be predicted; but the histogram of the
R
G
B
22
65
91
00010110
01000001
01011011
00011011
01000001
01011011
27
65
91
-Select lower value data channel among R, G and B
-Decide the maximum number of data bits to store
-Store the data-bits in that channel
IJSER
Intern ational Journ al of Scientific & E ngineering Research, Volume 7, Issue 3, March-2016 867
ISSN 2229-5518
IJSER © 2016
http://www.ijser.org
images which is shown in Fig. 7 shows a minor different in the
value of the components: R, G and B.
(a) Original carrier
(b) Carrier with secrete using modified algorithm
Fig. 6: Image Steganography testing example
Theoretically, the average number of bits are used per pixel
is equal to 3.62 where the average number for Triple-A is 3.42,
SCC is 3, and for LSB is 1. This shows us that the capacity of
the new proposed technique is higher than the previous tech-
niques. The average capacity ratio of our method is 15.08% of
the original cover media size. This is better than SCC and Tri-
ple-A algorithm where the capacity ratio is 12.50% and 14.28%
respectively. Another advantage of this technique is that the
use of minimum number of pixels to hide a message M inside
carrier C.
Tabl e 3: Comparing SCC, Triple-A and proposed technique
Size of M
(bytes) 28 KB
Average no.
of bits used
per pixel
Pixels used to
hide M inside C
Capacity
ratio
Our Proposed
3.62
6958
15.08%
SCC
3.00
27984
12.50%
Triple-A
3.43
7169
14.28%
Graph 1. Average number of bits and Capacity in KB
The main advantage of our proposed method is if we utilize
the color intensity value of cover-image to hide the data bits,
then our algorithm has very high capacity of data hiding as
compare Triple-A algorithm. In Triple-A [12] it does not utilize
the lower intensity based pixel for high capacity data embed-
ding. Table 3 and Graph-1 shows a comparison result of our
proposed algorithm with Triple-A and SCC. The table and
graph shows that the capacity ratio of our proposed method is
more than Triple-A. The result is obtained using different car-
rier images and averaging the number of pixels used in the
hiding operation. It also shows that our technique enhance the
capacity ratio without affecting the image with noise or distor-
tion.
(a) Original carrier
IJSER
Intern ational Journ al of Scientific & E ngineering Research, Volume 7, Issue 3, March-2016 868
ISSN 2229-5518
IJSER © 2016
http://www.ijser.org
(b) Carrier with secrete using SCC algorithm
(c) Carrier with secrete using Triple-A algorithm
(d) Carrier with secrete using Proposed algorithm
Fig. 7: Image Steganography histograms
5 CONCLUSION
In this paper we have explored the existing image Steganog-
raphy techniques, and have improved the data hiding capacity
of existing Triple-A algorithm [12]. Mainly the proposed algo-
rithm uses actual color of the channel in conjunction with
pseudo random number generator (PRNG) to decide the
number of data bits to store. In our technique we have utilized
the lower intensity based pixel for high data embedding. This
approach leads to very high capacity with visual quality is as
close to Triple-A algorithm.
ACKNOWLEDGMENT
The authors would like to express their sincere thanks to the
anonymous reviewers for their constructive feedback, which
helped significantly improving technical quality of this paper.
REFERENCES
[1] Adnan Gutub, Mahmo ud Ankeer, Muhammad Abu-Ghalioun, Ab-
dulrahmanShaheen, and AleemAlvi, “Pixel Indicator high capacity
Technique for RGB image Based Steganography”, WoSPA 2008 –
5thIEEE International Workshop on Signal Processing and its Appli-
cations, University of Sharjah, Sharjah, U.A.E. 18 – 20 March 2008.
[2] Adnan Gutub, LahouariGhouti, Alaaeldin Amin, TalalAlkharobi, and
Mohammad K. Ibrahim, “Utilizing Extension Character ‘Kashida’
With Pointed Letters For Arabic Text Digital Watermarking”, Interna-
tionalConference on Security and Cryptography - SECRYPT, Barce-
lona, Spain, July 28 - 31, 2007.
[3] Adnan Gutub and ManalFattani, “A Novel Arabic Text Steganogra-
phy Method Using Letter Points and Extensions”, WASET Interna-
tionalConference on Computer, Information and Systems Science
andEngineering (ICCISSE), Vienna, Austria, May 25-27, 2007.
[4] Mohammad TanvirParvez and Adnan Gutub, “RGB Intensity Based
Variable-Bits Image Steganography”, APSCC 2008 – Proceedings of
3rdIEEE Asia-Pacific Services Computing Conference, Yilan, Taiwan,
9-12 December 2008.
[5] N.F. Johnson, S. Jajodia, "Exploring Steganography: Seeing the Un-
seen", IEEE computer, Vol. 31, No. 2, pages 26-34, February 1998.
[6] Fridrich, Jiri. "Applications of data hiding in digital images." Signal
Processing and Its Applications, 1999. ISSPA'99. Proceedings of the
Fifth International Symposium on. Vol. 1. IEEE, 1999.
[7] F.A.P. Petitcolas, “Intro ductio n to information hiding”, in: S. Katzen-
beisser and F.A.P. Petitcolas, (ed.) (2000) Information hiding tech-
niques for steganography and digital watermarking, Norwood: Ar-
techHouse, INC.
[8] Provos, Niels, and Peter Honeyman. "Hide and seek: An introduction
to steganography." Security & Privacy, IEEE 1.3 (2003): 32-44.
[9] Silman, Joshua. "Steganography and steganalysis: an overview." Sans
Institute 3 (2001): 61-76.
[10] C.K. Chan, L.M. Chen, Hiding data in images by simple LSB substi-
tution, Pattern Recognition 37 (3) (2004) 469–474.
[11] S. Venkatraman, A. Abraham, M. Paprzycki, "Significance of Ste-
ganography on Data Security", International Conference on Infor-
mation Techno logy: Coding and Computing (ITCC'04), 5-7 April
2004.
[12] Gutub, Adn an, Aye d Al-Qahtani, and AbdulazizTabakh. "Triple-A:
Secure RGB image steganography based on randomization." Com-
puter Systems and Applications, 2009.AICCSA 2009.IEEE/ACS Inter-
IJSER
Intern ational Journ al of Scientific & E ngineering Research, Volume 7, Issue 3, March-2016 869
ISSN 2229-5518
IJSER © 2016
http://www.ijser.org
national Conference on.IEEE, 2009.
[13] Kevin Curran, Karen Bailey, “An Evaluat ion of Image Based Ste-
ganography Methods”, Multimedia Tools and Applications, Vol. 30,
No. 1, Pages: 55 – 88, July 2006.
[14] J.-L. Dugelay and S. Roche, “A survey of current watermarking tech-
niques,” in Information Hiding Techniques for Steganography and
Digital Watermarking, S. Katzenbe isser and F. A. P. Petitcolas, Eds.
Norwood, MA: Artech House, 1999, ch. 6.
[15] Bender, Walter, et al. "Techniques for data hiding." IBM systems jour-
nal 35.3.4 (1996): 313-336.
[16] Donovan Artz, Los Alamos National Laboratory, “Digital Steganog-
raphy: Hiding Data within Data”, IEEE Internet Computing: May,
2001.
[17] Nabarun Bagchi. Article: Secure BMP Image SteganographyUsing
Dual Security Model (I.D.E.A, image intensity andBit Randomiza-
tion) and Max-Bit Algor ithm. I nternationalJournal of Computer Ap-
plications 1(21)(2010):18–22.
IJSER
