www.ijraset.com Vol.

2 Issue IV, April 2014

ISSN: 2321-9653

INTERNATIONAL JOURNAL FOR RESEARCH IN AP PLIED SCIENCE

AN D E N G I N E E R I N G T E C H N O L O G Y (I J R A S E T)

Detecting Phishing Websites Based on Visual

Cryptography
Ambre Reshma1, Fugare Saroj2, Revagade Ranjana3 ,Wakchuare Priyanka4 , Deshmukh S.C. 5

Abstract/Most of the time people using internet for online transactions from one place to another place. The security in this case
should be very high and should not be easily traceable to attacker. Attacker may attack on system online or offline. In this types of
various attacks, phishing (meaning is- knowing users details through fake sites) is identified as a major security threat and each
second new ideas comes into place to attack users account online, soprevention mechanism should be very strong. In Phishing process
individual or a group of thefts used to steal person’s confidential information such as passwords, credit card information, banking
details etc.for making frauds. In this project, new approach is defined named as "A Novel Anti-phishing framework based on visual
cryptography" to solve phishing based problems. In project, an image based authentication using Visual Cryptography (VC) is used.
The privacy of image captcha is preserve by the use of visual cryptography by extracting the original image captcha into two shares
that are stored in separate database servers such that the original image captcha can be identified only when both are simultaneously
available.

Keywords/ Phishing, Visual Cryptography, Image Captcha, Shares, Security.

cryptography is used in our project. Image Processing is a
I. INTRODUCTION
technique of processing an input image and to get the
Most online applications use various types of securities. output as either improved form of the same image. In
Since, the design and technology of has improved rapidly, Visual Cryptography (VC)an image captcha is decomposed
hacking and cracking techniques also improved in same
range. As a result, it is nearly impossible to make sure into shares and stores separately in server database, so after
whether a computer that is connected to the internet canbe login by user, that image captcha should get match with its
trustworthy and secure. Phishing scams is become a content behind image. After both user and server side
problem for online banking and e-commerce users. Big captcha’s get matched, the user can log in into the website
question is how to prevent e transaction applications online very securely.
from Phishes. Phishes means the online theft that aims to
steal sensitive information such as online Banking II. RELATED WORKS
passwords and credit card information from users. Another Phishing web pages are forged web pages that are
meaning of phishing states that it is “the act of sending an created by malicious people to mimic Web pages of
email to a user claiming that they have won $100000000 original web sites. M These types of web pages have high
prize something or of any kind of fake advertising mails, so visual similarities to scam their casualty’s. Such types of
that to get users after clicking on it may indirectly give his web pages stare absolutely like the actual a solitary.
personal information to fraud users(theft)”. Casualties of fraud web pages may expose their personal
So in our project we introduce a new concept which account, secure word, positive identification number, or
can be used as a safe way against phishing which is named other data to the phishing web page holders. It includes
as "A novel approach against Anti-phishing using visual methods such as defraud the customers through email and
cryptography". In this methodology, website we can check unsolicited mail, fitting of key loggers. Electronic mails are
side by side verifies its own identity and proves that it is a mostly use techniques for phishing, due to its simplicity,
true website before the end users and makes sure that both ease of use and wide reach. Phishes can deliver specially
user and server sides are authenticated one. The technique crafted emails to millions of legitimate email addresses
of both image processing and an improved visual very quickly and can fool the recipients utilising well

Page 115
www.ijraset.com Vol. 2 Issue IV, April 2014

ISSN: 2321-9653

INTERNATIONAL JOURNAL FOR RESEARCH IN AP PLIED SCIENCE

AN D E N G I N E E R I N G T E C H N O L O G Y (I J R A S E T)
known imperfection in the Synchronous Mail Transfer blacklist has a lengthy delay time, the correctness of
Protocol. Methods like adenovirus extension to electronic blacklist is low.
mails, crafting of ‘personalised’ or unique email messages 2. Heuristic-based anti-phishing technique, with a high
are also often. Investigator proposes customer-based possibility of false and failed reminder, and it is not
mechanisms to authenticate the server. Automated difficult for the attacker to use technical means to avoid the
Challenge Response Method is one such authentication heuristic features diagnose.
mechanisms includes challenge generation module from 3.Similarity assessment based technique is time-
server which in turn interacts with Challenge-Response swallowing. It needs too more time to deliberate a couple
interface in client and request for response from user. Here of pages, so using the method to diagnose fraud websites
instead of getting response from get-response executable it on the client terminal is not fitting. And there is low
is better to update the get-response executable correctness rate for this method depends on many factors,
automatically from bank server when the responses are such as the word, photographs, and equivalence
about to negate. Now there are Domain Name System measurement technique. However, this technique is not
based anti-phishing resemble technique which mainly perfect enough yet.
includes blacklists, heuristic detection, the page similarity 1. 0ne time password: - If user forgot his/her
assessment. But they do have some short comings. password then user create new password using OLTP
Blacklist is a DNS based anti-phishing approach technique technique.
now most commonly used by the browser. Anti Phishing 2. Using visual cryptography scheme we divide
Work Group, Google and other organizations have image into six shares because of these when any share
provided an open blacklist enquiry interface. Internet get crash or lost from other, we recover the original
Explorer 7, Netscape Browser 8.1, Google Safe Browsing image.
(a feature of the Google Toolbar for Firefox) are important
browsers which use blacklists to protect users when they III. PHASES OF SYSTEM
are navigating through phishing sites. Because A. Registration Phase:
administrator has verified every URL in the blacklist, the
fault reminder possibility is very low. Although, there are a
plenty of technical drawbacks. Initially, the fraud websites
we found is a very minor percentage, so the failed reminder End User Server Enter
possibility is very big. Furthermore, commonly to say, the Enter Password
life revolution of a fraud website is not many days. A Password
website might be shut down before we found and verified it
is a fraud website. Heuristic-based anti-phishing technique
is to estimate whether a page has some phishing fact-
finding features. For example, some fraud-finding features Image Captcha
used by the Spoof Guard [3] toolbar include checking the Produced
host name, checking the Uniform Resource Locator for
common spoofing techniques, and checking against before
watch images. While you only use the Heuristic-based
technique, the perfection is not sufficient. Alongside,
frailer can use some policies to circumvent such detection First Share Second Share
rules. The user may be defraud by the fraud website Kept With End Kept With
because the phishing website imitates a legitimate website. user Server
Its pages are often similar with the legitimate sites. For Fig 1: Registration Phase
example, CANTINA is a content similarity based approach In registration phase, user enter password at the time of
to detect phishing websites [6]. These favored technologies registration for secure website. The password is
have several downsides: combination of numbers and alphabets. The password
1. Blacklist-based technique with low false reminder joining with the randomly generated password in the server
possibility, but it cannot diagnose the websites that are not side and image captcha produced. The image captcha
in the blacklist database. Because the life revolution of distinct into two shares one share kept with user & other
fraud websites is too short and the establishment of share kept with server. The user’s share & initial image

Page 116
www.ijraset.com Vol. 2 Issue IV, April 2014

ISSN: 2321-9653

INTERNATIONAL JOURNAL FOR RESEARCH IN AP PLIED SCIENCE

AN D E N G I N E E R I N G T E C H N O L O G Y (I J R A S E T)
Captcha is sent to the user during login phases. The image of the shares are overlaid the secret image is disclosed. The
captcha stocked in the real database of secure websites. user will be prompted for i, the number of participants.
After registration user can modify the password when it is 3. (i,i) Threshold VCS scheme-This scheme encrypts the
needed [6]. secret image to i shares such that when all i of the shares
are combined will the secret image be disclosed. The user
will be prompted for i, the number of participants.
B. Login Phase
4.(k,i) Threshold VCS scheme- This scheme encrypts the
In this phase’s user id then user entered share which is secret image to i shares such that when any group of at
kept with him. User Share send to the server the user share least k shares are overlaid the secret image will be
and server share are combined together. The image captcha disclosed. The user will be prompted for k, the Threshold,,
seen to the user .Here user can test seen image Captcha and i, the number of participants.
match with Captcha created at the time of registration. User
enters the text seen in the image captcha this text can used In that case (2, 2) VCS, each pixel P in the real image is
for password purpose and user can login to website.Using encrypted into two sub pixels called shares. Figure show
user id Image captcha produced by combining two shares the shares of a white pixel and a black pixel. Neither share
one can verify whether the website is secure website [6]. provides any clue about the original pixel since different
pixels in the genuine image will be encrypted using
User enter Share from independent random choices. When the two shares are
username, share server superimposed, the value of the real pixel value of p found.
The p black pixel then we gets two black sub pixels and if
p is white pixel, we get black sub pixel and another is white
sub pixel [6].
Captcha
Generated

Go to Wrong Phishin
home page Captch g Site
a

Fig 2: Login Phase

IV. VISUAL CRYPTOGRAPHY SCHEME

VC scheme is a cryptographic technique that allows for
the encryption of visual information such that decryption
can be performed using the human visual structure. We can
accomplish this information by following scheme.
1.(2,2) Threshold VCS scheme- This is a simplest
threshold scheme that takes a secret message and encrypts
it in two different shares that reveal the secret image when
they are overlap. No extra information is needed to create Fig. 3: VCS scheme with 2 sub pixel construction.
this kind of access structure.
V. OUR APPROACH TO DETECTING PHISHING
2. (2,i) Threshold VCS scheme-This scheme encrypts the WEBSITE
secret image into i shares such that when any two(or more)

Page 117
www.ijraset.com Vol. 2 Issue IV, April 2014

ISSN: 2321-9653

INTERNATIONAL JOURNAL FOR RESEARCH IN AP PLIED SCIENCE

AN D E N G I N E E R I N G T E C H N O L O G Y (I J R A S E T)
Architecture stacking of shares of two image captcha's resulting in
original captcha. In Case3 share1 of first image captcha is
combined with share2 of second captcha resulting in
User id Browse First unrecognizable form of Captcha [6].
User Share Case 1:

Display Produce Concatenate

Image Captcha user share and
server share
Case 2:

Diagnose whether the site is phishing

or not

Case 3:
Enter string which Phishing
is produce in image Website
captcha

Fig 4: User log into website

In this methodology, end user enter user name user id
and user browse for the share which is kept with user and Figure 4: Creation and concatenation of shares
send to the server. User’s share is stacked with server share It is observed that both original and reconstructed
and image captcha generated. Detect whether websites is image captcha's are related with high degree of correlation.
phishing site or not according to image captcha. If image The correlation coefficient of original captcha and
captcha matches with captcha generated at time of reconstructed captcha are shown in Table 1.Also when two
registration then user is human user and user enter the different shares are stacked their corresponding correlation
string displayed image captcha. Else Image captcha is not coefficient is obtained as -0.0073.This shows that there will
matches with captcha generated then at time of registration be zero degree of correlation between original and output
indentify phishing site[6]. images for two different shares [6].
VI. IMPLEMENTATION & ANALYSIS
In the registration phase the weightiest portion is the
make of shares from the image captcha .where one share is
kept with the user and other share can be kept with the
server side. For login phase, the user needs to enter a
correct username in the given text field. Then he has to
browse his share and process. At the server side the user's
share is joined with the share in the server and an image
captcha is generated .The user has to enter the text from the
image captcha in the required field in order to login into the
website. The entire process is depicted in Figure.5 as
different cases.Case1 and Case 2 illustrates the creation and

Page 118
www.ijraset.com Vol. 2 Issue IV, April 2014

ISSN: 2321-9653

INTERNATIONAL JOURNAL FOR RESEARCH IN AP PLIED SCIENCE

AN D E N G I N E E R I N G T E C H N O L O G Y (I J R A S E T)
VII. FUTURE WORK
1. This technology is not only use in banking sector
but also use in military sector for detecting
phishing websites
2. We will divide user confidential information onto
the five servers so it provides more security.
3. If user forgot his/her password so to create new
password we use new method such as one time
password.
4. We use new technique such as admin
authentication

VIII. CONCLUSION
Our methodology “Antiphishing framework based on
visual cryptography” preserves secure information of users
using three layers of security. First layer verifies whether
website is secure website or not. In second layer check
validated image captcha .Only human user acquire the
website can read image Captcha. In third layer of security
to prevent attacks from unauthorized user in user account.

REFERENCES
[1] Mrs.A.Angel Freda, M.Sindhuja, K.Sujitha, IJREAT,
Image Captcha based Authentication using visual
cryptography vol.1, pp.2320-8791, April-May, 2013.
[2] Anthony Y. Fu, Liu Wenyin, "Detecting Phishing
Web Pages with Visual Similarity Assessment Based
on Earth Moverâ Distance (EMD)",IEEE
Transactions on Dependable and Secure Computing,
vol.3, pp.301-311, October/December 2006.
[3] JungMin Kang, DoHoon Lee,IEEE, “Advanced White
List Approach for Preventing Access to Phishing
Sites”, pp.491-496, 2007.
[4] Wenyin Liu, Xiaotie Deng, Guanglin Huang, and
Anthony Y. Fu,IEEE Internet Computing "An
Antiphishing Strategy Based on Visual Similarity
Assessment", vol.10, p 58-65, March/April 2006..
[5] Haijun Zhang, Gang Liu, and TommyW. S. Chow,
IEEE Trans. “Neural Netw, Textual and Visual
Content-Based Anti-Phishing: A Bayesian
Approach”, IEEE Trans. Neural Netw., vol.22, no.
10, pp.15321546, Oct.2011.
[6] Divya James1 and Mintu Phillip2,”A novel anti
phishing framework based on visual cryptography”,
vol.3, no...1. pp.1264-3459, January.

Page 119