Towards Webpage Steganography with Attribute Truth Table

2016 3rd International Conference on Advanced Computing and Communication Systems (ICACCS -2016), Jan. 22 – 23, 2016, Coimbatore, INDIA
Towards Web Page Steganography with Attribute
Truth Table
B. SREEKANTH REDDY
M.Tech (Network & Internet
Engineering)
Dept. of Computer Science
Pondicherry University
Pondicherry, India, 605014
sreekanthbooch@gmail.com
K.S. KUPPUSAMY
Assistant Professor
kskuppu@gmail.com
T. SIVAKUMAR
Assistant Professor
tsivakumar@yahoo.com
Abstract— The evolution of steganography to use web pages
as cover medium has raised challenging research questions. Web
page information hiding is a way to conceal information apart
from its privacy policies like confidentiality, reliability and
availability. This paper proposes a Tag repository and attributes
truth table based method to hide data into web pages. The Tag
repository collects all potential tags and Attribute Truth Table is
built over the Tag repository for giving the binary values for the
attribute orders in the Tag repository. This Attribute Truth
Table helps in concealing the secrete information into the web
page. The decoding of the secrete information is carried out using
tailor made client side plug-in.
Keywords— Web page; information hiding; tags; tag
repository, attribute truth table;
I. INTRODUCTION
Steganography is a Greek word, which indicates covered
writing and essentially means to conceal in plain sight.
Steganography is a technique of concealing data within cover
medium. The motto of steganography is to conceal information
in such a way that the unintended recipients cannot suspect the
cover medium[1]. Most steganography techniques have been
used various multimedia files as cover media like text, image,
audio or video.
In the beginning days of digital steganography,
images were mostly used as cover medium later it was
improved by moving towards other digital media format such
as video, audio, and text file. In recent years Web pages have
become the new means of cover medium for hiding the secret
information[2]. The multimedia files are only used as the cover
medium but not as a transmission medium. Whereas in web
page information hiding process, the web pages can serve as
both the cover medium and transmission method to conceal the
secret information.
Web pages considers the source code i.e., markup tags of a
HTML file as a plain text, through which it will instruct a web
browser how to display[3]. As the HTML is case insensitive
we can write the tags and attributes in lower and upper cases,
changing the quotation marks of the attributes. This method
allows to embed secret information by switching the case of
letter in tag by considering lowercase letters indicates 0 and
uppercase letters indicates 1[4]. Apart from the tags of the
markup language we have another way of hiding the
information secretly is by changing the quotation marks of the
attributes (from single quotes to double quotes and vice-versa).
In this paper we have proposed a new approach of web
page information hiding by changing the order and case of
attributes in a combinatorial basis. Unlike existing approaches
in web page information hiding i.e. changing the case of tags
which can be done by changing some letters in the tags, this
paper is changing case for entire attribute so that an unintended
receivers cannot suspect it as a stego web page.
II. RELATED WORK
Steganography and Cryptography are used for data
confidentiality, but the main difference between the both is, in
cryptography process an intruder can know that both parties are
communicating in secret. Whereas steganography conceals the
existence of a secret information and nobody can suspect that
both parties are communicating in secret. This makes
steganography different from the cryptography technique[5].
In the existing models of the web page information
concealing methods the researches have been done to conceal
the secret information in the tags and attributes of the source
code i.e. HTML and CSS files [6] as well as the white spaces
of the source code [7][8]. The popular techniques used in the
existing model are as listed below:
A. Empty tag method:
In this process empty tags i.e. either a begin tag quickly
taken after by an end tag or an empty tag is used in order to
conceal the secret information. Using these types of empty tags
in the source code does not affect the content on the
webpage[9].
[978-1-4673-9206-8/16/$31.00 ©2016 IEEE]

B. Line break approach:
In this approach by continuously adding the line break tag
at the end of each tag the secret information was concealed in
the webpage. Usage of repetitive of line break at the end of
each tag does not affect the content of the web page[9].
C. Changing the case of the tag
As the HTML is not a case sensitive language even the
changes in the case of the tags may not show any change in the
web page while parsing[10][11].
D. Information hiding based on the attribute value string.
The strings used in the attribute values are not case
sensitive, so attribute values can be taken for information
hiding by keeping the uppercase indicates 1 and lower case
indicates 0 [2].
In this paper we are proposing an attribute casing and
attribute position reordering approach for concealing the secret
information. Later at the flip side which can be retrieved by
using the plug-in that works in reverse approach of hiding
algorithm.
III. PROPOSED MODEL
In the proposed scheme we are using the attributes of the
html for concealing the secret information, by changing the
order of the attributes and case of the attributes. In this
approach we are changing the order of the attributes and in
each order we are changing the case of the attributes. So we
will get huge number of attribute combinations. This will give
more possibility for concealing the secret information.
The proposed scheme is explained in the following way:
A. Secret information concealing procedure
1) Identification of potential tags:
It is impossible for concealing the secret information in
every HTML tags, because the tags such as line break can have
very less possibility for concealing information. So we have to
pick tags with some good number of attributes list, so that we
can conceal much information.
Such tags include Body, Font, Horizontal Ruler, Image,
Table, Table Row, and Table Definition.
TABLE I. ATTRIBUTE LIST
Tag Name Attribute List
Body Text , link , alink, vlink , bgcolor,
background
Font color , face , size
Horizontal
Ruler
Align, color, size , width
Image Align, alt , border , height , width , hspace
, vspace, src
Table Align, background, bgcolor, bordercolor,
cellpadding, cellspacing, cols, width
Table row Align, valign, bgcolor
Table
definition
Align, valign, background, bgcolor,
colspan, rowspan, width
If a tag has n-attributes
• We have n! Ways of changing the order of attributes.
• We have 2n
ways of changing the case of the attributes.
Here we are utilizing blend of both the aforementioned
methods i.e. firstly we change the order of the attributes and
then in each order we are changing the case of the attributes so
we will get (n!*2n
) combinations for each tag.
For instance consider the Font tag with 3 attributes size,
face, color then we will get the combinations as follows.
Order of the attributes:
[size, face, color], [size, color, face], [face, size, color],
[face, color, size], [color, size, face], [color, face, size]
With the first possibility, we have 8 ways to change the
case of the tags as follows:
TABLE II. ATTRIBUTE CASING
Attributes Bit
stream


000
 
001


010


100
 
110
<font SIZE="3" FACE="Aharoni"
111
This is for one order of attributes, like wise we will get
(n!*2n
) combinations for every tag with n attributes.
2) Tag Repository:
Tag Repository is a data base where we will save all the
possible forms of the attribute combinations. While saving the
tags in the tag repository we are arranging the attributes in the
alphabetical order.
3) Attribute Truth Table:
The Attribute Truth Table will take the tags from the Tag
Repository and assign the binary values for each of the tag.

With the goal that we have conceivable combinations and the
distinctive binary number for that attributes.
4) Information hiding process:
The secret information is converted into the binary stream
and the binary stream is concealed in the webpage using the
Attribute Truth Table i.e. a byte of binary stream is compared
with the Attribute Truth Table and corresponding attribute
settings are used for concealing the secret information.
5) Retrieval process:
The secret information which is concealed in the webpage
is retrieved using the plug-in shared by the web master. The
plug-in works in the bottom up approach of the hiding
algorithm.
The whole procedure is explained with the following
example:
We have designed a html webpage named Learn Easy for
online tutorial and we are concealing the secret information in
that webpage.
Secret information: Paris
We will insert a start of text and an end of text to indentify
the secret information
Decimal value of start of tag(STX) is 2
P = 112, a= 97, r=114, i=105, s=115
End of text (ETX) =3
Convert these ASCII values into binary stream.
The first tag we come across is <body> with 3 attributes so
we can have (3!*23
) values in the Attribute Truth Table, here
we need to conceal the binary value of “start of text”.
Next we come across the <table> tag with 5 attributes; here
we are hiding the P binary value, by changing the attribute
settings.
Next we have five <td> tags with 6 attributes each, here we
are hiding the binary values of a, r, i, s and the end of text.
This procedure is illustrated in the fig. 5
B. ALGORITHM
Algorithm: Hiding secrete message in web page
Input: Web page, Secret Information, Attribute Truth Table
Output: stego web page
1. Select the potential tags with more number of
attributes
2. Perform Attribute reordering as well as Attribute
casing
3. Store all these combinations in Tag repository
4. Create an Attribute Truth Table from Tag repository.
5. Convert the secret information in to the binary
stream.
6. Compare with the Attribute Truth Table and conceal
in the webpage.
7. Design a plug-in which works in the bottom-up
approach of the hiding process.
8. Share this with the intended user.
9. Stop
Fig. 1. Information hiding process
IV. EXPERIMENTAL ANALYSIS
When compared with the normal web page the stego web
page does not show any difference in viewing the content of
the web page. So this web page steganography is one of the
most advanced technology in steganography process, we had
experimented on a HTML web page and shown the results
below.
Fig. 2. Normal web page

Fig. 3. Source code of normal web page before hiding secret information
Fig. 4. Stego web page
Fig. 5. Source code of stego web page after hiding the secret information
The explanation is as follows:
<body bgcolor="white" TEXT="green" link="blue" >
Where the actual attribute order is set in the tag repository as
bgcolor, link, text. So we will have 48 possibilities in the
attribute truth table, since the value is 2 we will select
according to that.
<table BGCOLOR="yellow" border="2" cellpadding="10"
cellspacing="5" bordercolor="red" >
Where the actual attribute order of table is given as bgcolor,
border, bordercolor, cellpadding, cellspacing in the attribute
truth table.This has 5! Combinations and each combination
has 25
possibilities, as the value of P is 96 it will come in the
4th
combination and 17th
position as mentioned above.
<td ALIGN="left" bgcolor="white" colspan="1"
height="20%" width="10%" ROWSPAN="1" >Roll no.</td>
Where the actual attribute order of table definition is given
as align, width, height, rowspan, colspan, bgcolor in the
attribute truth table. This has 6! Combinations and each
combination has 26
possibilities, the value of a=97 comes in
34th
possibility in the 2nd
combination.
<td ALIGN="left" BGCOLOR="white" colspan="1"
height="20%" width="10%" rowspan="1" >NAME</td>
This has 6! Combinations and each combination have 26
possibilities, the value of r=114 come in 50th
possibility in the
2nd
combination.
<td ALIGN="left" bgcolor="white" colspan="1"
HEIGHT="20%"WIDTH="10%"ROWSPAN="1">COURSE<
/td>
possibilities, the value of i=105 comes in 39th
possibility in the
2nd
combination.
<td ALIGN="left" BGCOLOR="white" colspan="1"
height="20%”WIDTH="10%"ROWSPAN="1" MARKS</td>
possibilities, the value of s=115 come in 51th
possibility in the
2nd
combination.
<td align="left" bgcolor="white" colspan="1"
height="20%"ROWSPAN="1"WIDTH="10%” GRADE</td>
possibilities, the value of end of text= 3 come in 4th
possibility
in the 1nd
combination.
In this way we will conceal the secret information in the
webpage by changing the position of the attributes as well as
the case of the attributes.

V. CONCLUSION
In this paper we have proposed an attribute positioning and
attribute casing approach for concealing the secret information.
Contrast with the current webpage steganography techniques
which will works on the tag casing and attribute casing this
approach of changing the entire attribute name, which will
prevent the intruders not to suspect about the web page
steganography process. This approach give more possibility of
hiding the secret information because of the attribute truth
table. The secret information shall be accessed by the
authorized users with the specially client side plugin which
would utilize the attribute truth table to decode the hidden
information.
REFERENCES
[1] S. Channalli and A. Jadhav, “Steganography An Art of Hiding
Data,”International Journal on Computer Science and Engineering
vol. 1, no. 3, pp. 137–141, 2009.
[2] X. Yong, L. Juan, and Z. Yilai, “A High Capacity Information
Hiding Method for Webpage Based on Tag,” 2012 Third Int. Conf.
Digit. Manuf. Autom., pp. 62–65, 2012.
[3] Y. Yang and Y. Yang, “An efficient webpage information hiding
method based on tag attributes,” Proc. - 2010 7th Int. Conf. Fuzzy
Syst. Knowl. Discov. FSKD 2010, vol. 3, no. Fskd, pp. 1181–1184,
2010.
[4] D. Shen and H. Zhao, “A novel scheme of webpage information
hiding based on attributes,” Inf. Theory Inf. Secur. (ICITIS), 2010
IEEE Int. Conf., pp. 1147–1150, 2010.
[5] J. Ren and L. Zhang, “Webpage Information Hiding Algorithm
Based Webpage Information Hiding Model Based on the Tags,” pp.
67–74, 2012.
[6] “Steganography.” [Online]. Available:
https://en.wikipedia.org/wiki/Steganography.
[7] Y. C. Chou and H. C. Liao, “A Webpage Data Hiding Method by
Using Tag and CSS Attribute Setting,” 2014 Tenth Int. Conf. Intell.
Inf. Hiding Multimed. Signal Process., pp. 122–125, 2014.
[8] I. Lee and W. Tsai, “Secret Communication through Web Pages
Using Special Space Codes in HTML Files,” Int. J., pp. 141–149,
2008.
[9] D. V Dhawase and P. S. Chavan, “WEBPAGE INFORMATION
HIDING USING PAGE CONTENTS,” vol. 3, no. 1, pp. 182–186,
2014.
[10] X. Guo, G. Cheng, C. Zhu, A. Zhou, W. Pan, and D. Truong, “Make
Your Webpage Carry Abundant Secret Information Unawarely,”
2013 IEEE 10th Int. Conf. High Perform. Comput. Commun. 2013
IEEE Int. Conf. Embed. Ubiquitous Comput., pp. 541–548, 2013.
[11] S. Dey, H. Al-Qaheri, and S. Sanyal, “Embedding Secret Data in
Html Web Page,” arXiv Prepr. arXiv1004.0459, pp. 1–10, 2010.

Towards Webpage Steganography with Attribute Truth Table

More Related Content

What's hot (15)

Similar to Towards Webpage Steganography with Attribute Truth Table (20)

Recently uploaded (20)

Towards Webpage Steganography with Attribute Truth Table