![Line shifting where second line is shifted up 1/300 inch [7]](https://www.researchgate.net/profile/Ammar-Odeh/publication/256296824/figure/fig3/AS:667780764426243@1536222749230/Line-shifting-where-second-line-is-shifted-up-1-300-inch-7_Q320.jpg)
Content uploaded by Ammar Odeh
Author content
All content in this area was uploaded by Ammar Odeh
Content may be subject to copyright.
Content uploaded by Ammar Odeh
Author content
All content in this area was uploaded by Ammar Odeh
Content may be subject to copyright.
Steganography in Arabic Text Using Full Diacritics Text
Ammar Odeh
Computer Science & Engineering,
University of Bridgeport,
Bridgeport, CT06604, USA
Aodeh@bridgeport.edu
Khaled Elleithy
Computer Science & Engineering,
University of Bridgeport,
Bridgeport, CT06604, USA
elleithy@bridgeport.edu
Abstract
The need for secure communications has significantly
increased with the explosive growth of the internet and
mobile communications. The usage of text documents has
doubled several times over the past years especially with
mobile devices. In this paper we propose a new
Steganogaphy algorithm for Arabic text. The algorithm
employs some Arabic language characteristics, which
represent as small vowel letters. Arabic Diacritics is an
optional property for any text and usually is not popularly
used. Many algorithms tried to employ this property to hide
data in Arabic text. In our method, we use this property to
hide data and reduce the probability of suspicions hiding.
Our approach uses a performance metric involves the file
size before and after adding Diacritics and ability to hide
data with being suspicious.
Keywords: Kashida, Carrier file, Zero width chracter, Information
Hiding, Diacritics.
1. INTRODUCTION
1.1. Background
Steganogaphy is a combination of two Greek words. Where
"Stegano" means hidden and “Graptos" means writing. The
secure data is embedded into other object, so a middle
attacker can't catch it [1]. Invisible ink is an example for
Steganogaphy where a readable message is transferred
between source and destination. Any intruder in the middle
can read the message without knowing about the hidden
data. Meanwhile, authorized persons can read it depending
on substances features [2], [3].
Ancient Greece used to shave the messenger head and then
wait until hair grows back, then the message can be sent to
the destination [1]. Depending on this method, we have 2
possibilities:-
1. If the message arrives, the receiver can read the message
and recognize if the message changed or not.
2. If the message did not arrive, this mean an attacker was
able to detect the message.
1.2. Motivation
Steganogaphy algorithms depend on three techniques to
embed hidden data in carrier files.
1. Substitution: Exchange some small part of the carrier file
by hidden message. Where middle attacker can't observe the
changes in the carrier file. On the other hand, choosing
replacement process it is very important to avoid any
suspicion. This means to select insignificant part from the
file then replace it. For instance, if a carrier file is an image
(RGB) then the least significant bit (LSB) will be used as
exchange bit [4].
2. Injection: By adding hidden data into the carrier file,
where the file size will increase and this will increase the
probability being discovered. The main goal in this
approach is how to present techniques to add hidden data
and to void attacker suspicion [4].
Propagation: in this approach there is no need for cover
object. It depends on generation engine which is fed by
input (hidden data) to produce a mimic file (graphic or
music or text document).
Steganogaphy process consists of three main components as
show in Figure 1.
Figure 1. General components of Steganogaphy
Different types of cover media can be used in Steganogaphy
including image, sound, video and text. Choosing a carrier
file is very sensitive where it plays a key role to protect the
embedded message. Successful Steganogaphy depends on
avoiding suspicion. Stegoanalysis starts checking a
transmitted file if there is any suspicion and this defeats the
main goal of Steganogaphy [3][4].
Text Steganogaphy represents the most difficult type, where
lack of data redundancy in text file in comparison with other
carrier files [5], which reduces the capacity of the hidden
data. Furthermore, text Steganogaphy depends on language
used, where each language has some characteristics which is
completely different from others. For example, letter shape
in English language doesn’t depend on its position in the
word, where Persian/Arabic letters have different forms
depending on their position in the word [6].
The algorithm presented in this paper aims to hide text
inside text by employing Arabic language and applying a
random algorithm to distribute the hidden bits inside the
message. The main reasons for choosing the Arabic
language are:
1. The proposed algorithm depends on multi dotted points
letters. Therefore, the algorithm must employ a language
that has as much as possible of dotted letters. For example,
the Arabic Language has 5 multipoint letters; the Persian
language has 7 letters [7], where the English language does
not have any one.
2. Wealth availability of electronic textual information.
3. There is a little research on other languages compared to
English.
4. Can be extended to other languages like Urdu, Farsi
and Kurdish.
1.3. Main Contributions and Paper Organization
An efficient algorithm is presented in this paper. The main
idea is to use Kashida, Zero width characters in Arabic that
enable us to hide two bits per one letter. Most of the
pervious algorithms hide one bit for one letter. In addition
we will use parallel connection, randomization strategy to
avoid any suspicions.
The rest of this paper is organized as follows. In section II
we discuss some text Steganogaphy techniques. Employing
Kashida and Zero width hidden algorithms are discussed in
section III. Finally, conclusion remarks are offered in
section IV.
2. PRIOR WORK
Text Steganogaphy is divided into two categories. The first
one is semantic based while the second is formatting based
as shown in Table I. In this section we present some of these
examples. In Table I we present simple comparison between
semantic and formatting methods.
Table I. Comparing between texts Steganogaphy
Semantic Method
Format Method
Amount of hidden
data
Small amount
More than semantic
Flaws
Sentence meaning
notice from OCR or
retyping
A Steganogaphy evaluation criterion depends on the amount
of data that can be hidden and the main problem faces the
method.
In this section we compare ten algorithms to hide data inside
text documents. The last two of them deal with Arabic and
Persian languages.
1. Word Synonym
Word Synonym is classified as semantic method and it
depends on replacing some words by their synonym. This
technique will convey data without any suspicion.
Meanwhile, the hidden data is small relative to other
methods. Moreover it may change the sentence meaning
[7][10][11].
2. Punctuation method
Using punctuations like (.) and (;) to represent hidden text
for example, “NY, CT, and NJ" is similar to "NY, CT and
NJ" where the extra comma is used to represent 1 or to
represent 0. The amount of hidden data in this method is
very small in comparison to the amount of cover media.
Inconsistence use of punctuation will be noticeable from a
Stegoanalysis perspective [9].
Table II. Using Word Synonym
Word
Synonym
Big
Large
Find
Observe
Familiar
Popular
Dissertation
Thesis
Chilly
Cool
3. Line Shifting
Line shifting involves vertically shifting a line a little bit to
hide information to create a unique shape of the text.
Unfortunately line shifting can be detected by character
recognition programs. Moreover, retyping the document
will remove all hidden data. In Figure 2 we present some
example about line shifting where the vertical shifting will
be very small (1/300 inch) so in normal case no one can
notice it.
Figure 2. Line shifting where second line is shifted up 1/300
inch [7]
4. Word Shifting
In this method, changing space between words will enable
us to hide information. Word shifting will be noticeable by
Optical Character Recognition (OCR) by detecting space
sequence between words.
5. SMS Abbreviations
Recently most transmitted SMS messages are using
abbreviations for simplicity and secure communications in
different applications such as internet chatting, email, and
mobile messaging.
The main advantage of this method is to save the time of
writing and the space needed to write messages and
manipulated keyboard limitation character.
Certain algorithms use some numbers to convey specific
information. As mentioned above SMS abbreviations can be
used in specific applications. In case these abbreviations are
not used in the common applications, suspicion from
Stegoanalysis systems might be raised.
Table III. Some SMS Abbreviations
Abbreviation
Meaning
ADR
Address
ABT
About
URW
You are welcome
ILY
I love you
EOL
End of lecture
AYS
Are you serious?
6. Text Abbreviations
Text abbreviations are similar to SMS abbreviations, where
a dictionary is created between each word abbreviation and
its meaning. The abbreviations dictionary is published to the
communicating parties. Abbreviation represents one method
to hide data. For example if you send “see” that is meant to
be “do you understand”[12].
7. HTML Spam Text
This method depends on HTML pages, where their tags and
their members are insensitive. For example <BR> is equal
to <Br> , <br> and <bR>. The hidden data depends on
letter case upper or lower to embedded 0 or 1.
8. TeX ligatures
Some special groups of letters are joined together to create
single glyph as shown in figure 3. The algorithm finds the
availability ligature in the text to hide a bit in each one. For
example if we want to hide 1 we write fi to f {}i which
creates some space between f and i. Otherwise we encode 0
[5].
The same algorithm can be applied to Arabic character ""
or " ". This algorithm has two problems. The first one is
that the file size increases when we apply extension in our
text. The second problem is that if OCR notes the font
change, it can be easy recognize the hidden message [6][5].
9. Vertical displacement of the points
This algorithm performs well if it is applied in pointed
letter. The English language has only {i, j} as pointed letters
In contrast, the Arabic and Persian languages have pointed
character sets (Arabic has 26 letters which 13 of them are
pointed, Persian has 32 letters which 22 of them are
pointed). The algorithm encodes 1 to shift up the point
otherwise encode 0. This method can encode a huge number
of bits, and need a strong OCR to recognize the changes. In
contrast retyping will remove the entire message [7].
10. Arabic Diacritics
Arabic language uses different marks to distinguish between
words that have same letters. it depends on Arabic Diacritics
(Harakat), where Diacritics are optional.
Figure 3 .Join between characters [5]
Figure 4. Vertical shifting point [7]
Most of Arabic scripts can be read without Diacritics which
depends on language grammar. The most occurrence is
Fatha "
" which can be used to encode 1 otherwise
encode 0.This algorithm enhances the reuse of cover media.
Furthermore, the carrier file size is reduced depends on
hidden message. Meanwhile, ORC scans the message with
different diacritics can conclude that there is hidden data, in
addition retyping will remove the embedded message [8].
Other research efforts studied adding extra diacritics to text
like [11], and to increase robustness of data used other
scenario to hide data in image.
In [17], the algorithm hides data by showing the diacritics if
it encodes 1, otherwise it hides the diacritics. The main
disadvantage of this approach that it might be observed if it
is compared more than once with the original text.
11. Using the Extension ‘Kashida’ Character
The strategy used in this method depends on letter extension
(Kashida). Kashida cannot be added at the beginning and at
end of word but it can be added between letters in words. In
other words, un-pointed letter with extension is used to hide
zero and pointed letter with extension will hold 1. [13] In
this approach, message content will not be affected. On
other hand, a new Unicode will be added (0640).
Table IV. Some Letters with mark and their Pronunciation
Pronunciation
Letter with
Haraka
Haraka
Do
Dama
De
Kasra
Da
Fatha
Figure 5. Kashida character after pointed letter [14]
As shown in Figure 5, not all characters can hide a bit.
Therefore, Stegoanalysis may suspect the message and
discover the embedded content.
12. Using the Extension ‘Kashida’ Character
This algorithm uses one Kashida to represent zero and 2
Kashida represent one. Since each letter needs 16 bits to be
represented, the algorithm use-mapping table to which each
character is mapped. So instead 16 bits, we can represent
each letter by 6 bits only and this will save 10 bits. [15]
13. Using Pseudo-Space and Pseudo Connection
characters
This technique is also called zero width non-joins (ZWNJ)
and zero width joiner (ZWJ) characters. The algorithm
classifies letters to join or non-join letters. If it is required to
hide 1 zero width is added, otherwise hide 0 is hided. [16]
3. PROPOSED ALGORITHM
Some of the Arabic characters features support different
Steganogaphy algorithms. In The following section, we
explore some Arabic languages properties to use some of its
attributes to hide large amount of data inside Unicode file
(Arabic language).
1. Writing Direction:-
Arabic text written from right to left. It is a unidirectional
language and numbers are read and written in the same
direction.
2. Letter connectivity:-
Most of the Arabic letters in the word are connected with
the previous letter and next one. Therefore, the letter may
have different shapes depending on its position in the word
as shown in Figure 6.
3. Dotted letters
Some Arabic letters have one, two and sometimes three
points. These points affects the letter’s pronunciation as
Table V shows.[18]
Figure 6. Different (Mem) Letter shapes depend in its position in the word.
4. Diacritic Signs (Harakat)
Arabic language has nine Diacritics as Table VI shows.
Usually those Diacritic optionally appears in most Arabic
text.
The main technique used in our algorithm is to employ
optional properties of Arabic language, which is the
Diacritic. We will be applying vertical shifting of Diacritic
relatively to the character. Zero is represented by no change
and 1 is represented by increasing the distance between the
letter and its dicraties as shown in Table VII.
Table V: Some pointed letter in Arabic letter
Table VI: Arabic Diacritics
Therefore, using this optional property in the text enables us
to pass a large amount of data by using small cover media
(text). Usually Diacritic does not have a standard distance or
position with respect to the letter Therefore, if we apply
Name of Diacritic
Shape of Diacritic
Dhammah
Sukkon
Fatha
Kasrah
Shaddah
Tanween Fatha
Tanween Dham
Mad
Tanween Kasr
vertical shifting by 1/200 inches it will not be noticeable by
any Stegoanalysis tools. Moreover, the text size will not be
largely affected.
Figure 7. Grid systems for none changing Arabic text
Figure 8. Grid systems for vertical shifting
Algorithm I Data Hidden
Read text from a website , Hide message from user
While !EOF // not end of file
Read letter and check
If letter has Diacritic then ==True
{// start if statement
Read bit from hidden message
If Hidden bit ==1 then
Shift up Diacritic
Else
Nochange
}/end if statement
Insert letter to Outputfile//stegotext
Next Letter
End While
Output Segotext
Algorithm II Data Extracting
Read text from Segotext
While !EOF // not end of file
Read letter and check
If letter has Diacritic then ==True
{// start if statement
If Shift up Diacritic == True
then
Insert 1 to output text
Else
Insert 1 to output text
}/end if statement
Next Letter
End While
Output Hidden data
4. DISCUSSION AND ANALYSIS
We applied our algorithm into eight websites as table XX
shows
In Table VIII we analyze the website by applying different
matrices.
1. File size:
Table VIII displays the size of the websites with and
without Diacritic and the difference between these two
cases. Most of websites show a change in their size less
than 1.4 KB and a ratio less than 7% of the website size.
2. Capacity ratio:
To represent the percent of the number of bits that can be
hidden in a specific file size, we use:
Capacity ratio =Number of bits can be hidden/ carrier file
size
Figure 9 shows relation of number of website data can be
hidden inside it.
Table VII Vertical shifting of Diacritic
Word
Hidden
Data
1
0
Figure 9 relations between different website and amount of data can be
hidden inside it.
Where data hidden inside page related to size of that page,
as figure 10 shows the difference between diacritic text and
none Diacritic text file size.
Figure 10 shows the size difference between diacritic text
and none Diacritic text
The results of the experiments performed illustrate that the
presented algorithm has the following advantages:
1. Size of cover media (text file) is minimally affected as
shown in Figure 10.
2. Large amount of data can be hidden inside the cover
media as Figure 9 shown
3. File format not effected in abnormal
4. About two billion people use Arabic language throughout
the world
5. Can be extended to other similar languages like Pashto,
Urdu, and Persian.
The main disadvantage of the presented algorithm is that
nowadays most of the Arabic text did not include diacritics.
Using diacritics may lead to some suspicions on carrier file.
5. Conclusion
Hiding data in different cover media represent one of
challenging security issues. One of the difficult media to use
for hiding data is a text, where embedding data may affect
the text format. The file size and format change will
increase the probability of being discovered using
Stegoanalysis tools and this will lead to reveal the hidden
data. The algorithm presented in this paper use ones of
Arabic language property, which is called diacritic, without
effect on the file size or text format in any abnormal or
suspicious way. Comparison of this algorithm with other
techniques in the same categories [8][11], our algorithm
does not remove any of the text diacritics and this will
reduce the probability suspicion by Stegoanalysis tools.
Furthermore, vertical shifting will not affect the file size in
any noticeable way.
0
500
1000
1500
2000
2500
3000
1
2
3
4
5
6
7
8
Number of Hidden bits
Web site
Web site and amount of hidden bits
0
5
10
15
20
1
2
3
4
5
6
7
8
Website size in KB
Web site
website size effect by adding dicrities and
remove them
Table VIII Vertical shifting of Diacritics
Website
s
Number
of letter
Number
of letter
without
Size
without
Diacritic
Number
of
Diacritic
Size
without
Diacritic
Size
with
Diacritic
change
in size
Size
Ratio
Capacit
y Ratio
(Bit/Kilo
byte)
1
http://mubashermisr.aljazeera.net/
6138
3631
2507
16.6
17.19
0.59
3.55%
145.84
2
http://www.alfikralarabi.org/
6492
3854
2638
17.4
17.9
0.5
2.87%
147.37
3
http://mentouri.ibda3.org/
5651
3363
2288
16.6
17.6
1
6.02%
130.00
4
http://alamatonline.net/
7342
4352
2990
17.9
19.1
1.2
6.70%
156.54
5
http://www.sawaleif.com
2382
1479
903
14.8
15.3
0.5
3.38%
59.02
6
http://www.ahram.org.eg/
4272
2444
1828
15.7
16.5
0.8
5.10%
110.79
7
http://www.alquds.co.uk/
1661
1006
655
14.3
14.8
0.5
3.50%
44.26
8
http://www.alriyadh.com/
6512
3951
2561
17.5
18.4
0.9
5.14%
139.18
6. References
[1] Aelphaeis Mangarae "Steganography FAQ," Zone-H.Org
March 18th 2006
[2] S. Dickman, "An Overview of Steganography," July 2007.
[3] V. Potdar, E. Chang. "Visibly Invisible: Ciphertext as a
Steganographic Carrier," Proceedings of the 4th International
Network Conference (INC2004), page(s):385–391, Plymouth,
U.K., July 6–9, 2004
[4] M. Al-Husainy "Image Steganography by Mapping Pixels to
Letters," 2009 Science Publications
[5] M. Shahreza, S. Shahreza, “Steganography in TeX
Documents,” Proceedings of Intelligent System and Knowledge
Engineering, ISKE 2008. 3rd International Conference, Nov. 2008
[6] M. S. Shahreza, M. H. Shahreza, “An Improved Version of
Persian/Arabic Text Steganography Using "La" Word”
Proceedings of IEEE 2008 6th National Conference on
Telecommunication Technologies.
[7]M. H. Shahreza, M. S. Shahreza, “A New Approach to
Persian/Arabic Text Steganography “ Proceedings of 5th
IEEE/ACIS International Conference on Computer and
Information Science 2006
[8] M. Aabed, S. Awaideh, A. Elshafei and A. Gutub “ARABIC
DIACRITICS BASED STEGANOGRAPHY” Proceedings of
IEEE International Conference on Signal Processing and
Communications (ICSPC 2007)
[9] W. Bender ,D. Gruhl ,N. Morimoto ,A. Lu “Techniques for
data Hiding” Proceedings OF IBM SYSTEMS JOURNAL, VOL
35, NOS 3&4, 1996
[10] K. Bennett, “Linguistic Steganography : survey, analysis, and
robustness concerns for hiding information in text” Center for
Education and Research in Information Assurance and
Security, Purdue University 2004
[11] M. Nosrati , R. Karimi and, M. Hariri ,” An introduction to
steganography methods” World Applied Programming, Vol (1), No
(3), August 2011. 191-195.
[12] M.H. Shirali-Shahreza, M. Shirali-Shahreza, " Text
Steganography in chat" Proceedings of 3rd IEEE/IFIP
International Conference in Central Asia on Sept. 2007
[13] Adnan Abdul-Aziz Gutub, Wael Al-Alwani, and Abdulelah
Bin Mahfoodh “ Improved Method of Arabic Text Steganography
Using the Extension „Kashida‟ Character” Bahria University
Journal of Information & Communication Technology Vol.3, Issue
1, December 2010
[14] A. Gutub, L. Ghouti, A. Amin, T. Alkharobi, M. Ibrahim.
“Utilizing Extension Character Kashida with Pointed Letters for
Arabic Text Digital Watermarking”. Proceedings of the
International Conference on Security and Cryptography,
Barcelona, Spain, July 28-13, 2007, SECRYPT is part of ICETE -
The International Joint Conference on e-Business and
Telecommunications. pages 329-332, INSTICC Press, 2007
[15] Adnan Abdul-Aziz Gutub, and Manal Mohammad Fattani,
“A Novel Arabic Text Steganography Method Using Letter Points
and Extensions “ World Academy of Science, Engineering and
Technology 27 200
[16] H. Shahreza, M. Shahreza "STEGANOGRAPHY IN
PERSIAN AND ARABIC UNICODE TEXTS USING PSEUDO-
SPACE AND PSEUDO CONNECTION CHARACTERS".
Journal of Theoretical and Applied Information Technology.
[17] M. Bensaad, M. Yagoubi "High Capacity Diacritics-based
Method For Information Hiding in Arabic Text" 2011
International Conference on Innovations in Information
Technology.
[18] A. Azmi and A. Alsaiari "Arabic Typography: A Survey"
International Journal of Electrical & Computer Sciences IJECS
Vol: 9 No: 10
