2nd Int'l Conf.

on Electrical Engineering and Infonnation & Communication Technology (ICEEICT) 2015

Jahangirnagar University, Dhaka-1342, Bangladesh, 21-23 May 2015

Design and Implementation of Iris Recognition

Based Attendance Management System
IAmena Khatun, 2A. K. M. Fazlul Haque, 3Sabbir Ahmed, 4Mohammad Mahfujur Rahman,
1,2,3,4 Department of Electrical and Telecommunication Engineering
Faculty of Engineering
Daffodil International University

Iamena.ete@diu.edu.bd,2akmfhaque@daffodilvasity.edu.bd, 3sabbir@diu.edu.bd,
4mahfuj.ete@daffodilvarsity.edu.bd

Abstract- In the entire globe any educational biometric based attendance management systems.
organization is concerned in relation to the attendance Existing attendance management system is so much
of individuals because this has an effect on their overall time consuming as it takes time to call students name
performances. In conventional method attendance of or signing on the paper. An additional problem of
students are taken by calling student names or signing having the hardcopy of the attendance record is that
on paper which is extremely time overwhelming. To the attendance sheet may misplace [2]. A biometric
eliminate this problem one of the solutions is a based automatic attendance management system was
biometric-based attendance system that can proposed that use fmgerprints for identification. For
automatically capture students' attendance by reliable recognition of persons an alternative
recognizing their iris. Iris recognition is regarded as one approach is iris pattern which is very effective
of the most reliable, accurate and efficient biometric
specially when need to search large databases without
identification system due to the inner characteristics of
any false matches [3]. A person can be reliably
iris, such as uniqueness, immovability and time
identified by iris recognition because of its
uniqueness and stability. It is becoming so popular
invariance. The aim of this paper is to design and
because no two irises are identical, even irises of the
implement an iris recognition based attendance
left and right eye of the same person are not alike.
management system with the latest facilities at an
Irises are also stable because one's iris is fully
accessible price to think about the financial situation of
structured by ten months of age and remains the same
the large figure of developing countries. The paper
for the period of their life span while other
includes two parts, hardware and software. The
identifying features can change with age [4]. The
hardware part is responsible to capture images and run
accuracy of iris recognition technology is more as it
required program whereas the software part is
uses 240 reference points of iris for matching. In
responsible for accusation, processing, iris localizing,
1936 Frank Burch, an ophthalmologist proposed the
adjusting, matching and storing data; send the
system to identify a person using iris pattern and the
attendance report to the predefined E-mail address. idea of identification a person by his iris pattern was
Finally, the system detects, calculates, stores, and first appeared in the film of James Bond [5]. Aran
transmits the results by performing the MATLAB Safrr and Leonard Flom, two other ophthalmologists
program. patented this idea in 1987 and in 1989 these two
ophthalmologists asked John Daugman to fonn
Keywords-Biometric; Iris Recognition; Radius algorithms for iris recognition. Iris extraction feature
Calculation; MATLAB; Authentication. provides exceptional high dimensional information
that clarifies the popularity of iris recognition based
I. INTRODUCTION identification system [6, 7]. The iris is a circular
diaphragm that punctured because of the pupil and
In today's world peoples are facing many problems lies between cornea and lens. How much light can
and fake identity is one of the core problems in the enter through the pupil, a circular aperture area is
whole world [1]. As move towards the digital era controlled by the iris. The pupil size can vary from
biometric technologies become more and more 10% to 80% of the iris diameter where the diameter
popular. Iris recognition is one of the most efficient is approximately 12mm [8]. Individual's iris pattern

978-1-4673-6676-2115/$31.00 m015 IEEE

will be used for attendance instead of signing an checking, iris extraction, iris matching, data storing,
attendance sheet, once a match is made he or she will authentication
be registered. The proposed system uses the camera
to capture live images of human iris and stored this
information in a data base. The proposed system
calculates the radius of iris using Gray-coding
algorithm in MATLAB data analyzing software. The
system also uses MATLAB to match the radius of
each person with the previous stored value and
fmally send the attendance report to the predefmed E­
mail address without human intervention.

II. PROPOSED SYSTEM

In future numerous business enterprises will be relied

on biometrics technologies since it is the only way to
provide guarantee about the presence of an individual
[9]. The proposed system provides iris recognition Figure 2: Web-cam with a Laptop.
based biometric system in where the web-cam takes
images of persons and sends to the computer for and to send E-mail. After capturing the image only
further processing and storing for future usages. The iris need to localize from the whole image. Iris can be
basic diagram of iris recognition system is illustrated localized from the edge color of the pupil that is used
in figure 1. The images are stored in data bases that in MATLAB coding to extract eye. MATLAB
will be used for comparing with the next images of extracts the appropriate images of eyes to get an
person for authentication. If the next image gets accurate value that will be stored in the database for
matched with the stored one then the message goes to authentication. The system uses different parameters
the administrative computer and if not matched then for running the program such as Gray coding, color
the previous process will run repeatedly. A webcam adjustment, and pixel adjustment. If given images
is a video camera typically connected by a USB are not in appropriate position then the calculation
(Universal Serial Bus) cable that sends the image to a will not be held and it will require again input images
computer or computer network. Computer is used for or reject the program. Only eye part is extracted from
data storing, processing and analyzing for iris snap short by MATLAB imcrop tool which uses
recognition process. some parameters such as position, line width, line
style, edge color to calculate the radius. Iris matching

�----.��---���.���
process is a comparing coding system that takes
images and compare with the store database. If this
process match then the attendance will transmit or it
will required the same process again.

The taken data and the calculation are stored in the

specified drive of the computer. When data are
needed then it calls the data for comparison.
Authentication is a process of providing access to a
system for security reason based on individual's
Figure 1: Basic diagram of hardware processing of
identity. It ensures who is the right person or who has
iris recognition of attendance management system.
the right to access. But the data identity of a person
has to store before in the database and when his or
After capturing and storing of images, MATLAB
her image is given then the input images will be
data analyzing software, is used for iris image
compared with stored image. If this process gets
acquisitions, iris localization, iris adjustments, iris
matched then it goes to the next steps as shown in the
flow chart (figure 3).
 Figure 5: Steps followed sending an E-mail through
the SMTP server

Figure 3: Simplified block diagram of the process. III. RESULT AND ANALYSIS

The system also sends the attendance report to A. IMAGE ACQUISITION AND RESIZING
predefmed E-mail address without human
Firstly, the system will ask for permission to take
intervention. To enable automated E-mail sending
images using the MATLAB code. After getting
system, one will have to access a portable SMTP
permission the system captures images. The captured
(Simple Mail Transfer Protocol) server, an internet
images are then stored for further analysis. After
standard offered by Gmail to send mail from any
capturing the image MATLAB imcrop tool detect
network to any E-mail address. To access it, one has
only the eye part from the whole image and calculate
to develop email client software. This task has done
the radius of iris.
in MATLAB using the built in functions. The
architecture of the system has shown in figure 4. For
this service the PC must be connected with the
internet, here citycell zoom modem is used for data
connectivity. The E-mail can be sent to the given E­
mail address through SMTP server.

Figure 4: Simplified block diagram for implementing

Automated E-mail Service

Figure 6: Image capturing window.

The step of sending report to the E-mail address are
shown in figure 5.
 Figure 7: Automatically cropped part of eye Figure 8: Pixel of iris

C. AUTHENTICATION

B. RADIUS CALCULATION Authentication is the method of ensuring the identity

of an individual. Authentication plays a significant
Angular Radial Partitioning (ARP) method is used to role which can provide the assurance of person's
compute the radius of iris from the cropped part of whether he or she has the right to access. In this
eye. The system automatically calculates data of proposed system after processing and calculating
radius which will be stored in the database radius of the image the result are stored in the data
simultaneously. For the first person the radius of iris bases that will be used for comparing with the next
is 159.625 pixels as shown in MATLAB window in person's image for authentication. If the next image
Figure 8. is matched with the stored one then the system shows
that the value is matched with stored value and takes
this person's attendance.

Figure 7: Radius of iris in MATLAB command

Figure 9: Authentication message
window
If the value of radius is not matched with the E. Accuracy
previous stored value on the database, the system
shows that the value is matched with stored value so
the process will be continued.

Figure 12: Result of the system

Figure 10: Not Authentication message.
The result of the system, illustrated in figure 12
D. AUTOMATED E-MAIL SERVICE presents the radius values of iris of a same person in
different times. So the accuracy can be determined by
After completion of authentication the report will be the following way.
sent to the predefined E-mail address. In the
MATLAB coding the E-mail address has already Testl-Test2= (135.7315-135.4277) = 0.3031
been defined. The PC must be connected by the
internet for this purpose. In this work city cell zoom Testl-Test3= (135.7315-135.5213) = 0.2102
modem is used for internet connection in the system
Testl-Test4= (135.7315-135.7101) = 0.0214
that take about 3 seconds to automatically send the
attendance report with some information to the
= 0.5347/3
predefined E-mail address after the completion of = 0.1782x100
acquisition and post processing. = 17.8%
Accuracy is = (100 - 17.8) % = 82.2%

After analyzing, the error and accuracy can be

calculated. The values of iris radius of an individual
have taken several times, comparing these values
with the stored value of this person error and
accuracy is 17.8% and 82.2% respectively.

IV. CONCLUSION

Now-a-days attendance management system plays an

important role for many organizations. The paper has
Figure 11: Automated E-mail
presented an automated attendance technique in
which an iris recognition system is used. This system
possesses the functions of iris image acquisitions, iris
localization, iris adjustments, iris checking, iris
extraction, iris matching, data storing, authentication
and sending E-mail to the pre-defmed E-mail address
without human intervention. The performance of this
system meets the needs of daily attendance
management in various enterprises and institutions.
By exploiting the benefit of advanced data analyzing
techniques, the cost of basic hardware unit was
successfully reduced. The proposed outcomes will
help one to go for a low cost attendance system just
using a pc, web-cam and wireless server. This work
can initiate a new era of attendance management
system.

REFERENCES

[I] Gajendra Singh Chandel, Ankesh Bhargava , "Identification

of People by Iris Recognition", International Journal of
Science and Modem Engineering (IJISME) ISSN: 2319-
6386, Volume-I, Issue-4, March 2013.
[2] M. Mattam, S. R. M. Karumuri, and S. R. Meda,
"Architecture for Automated Student Attendance," in Proc.
IEEE Fourth International Conference on Technology for
Education (T4E 2012), pp.164-167, 18-20 July 2012, doi:
10.1109/T4E.2012.39.
[3] Seifedine Kadry, Khaled Smaili, "A design and
implementation of a wireless iris recognition attendance
management system", ISSN 1392 - I24X INFORMATION
TECHNOLOGY AND CONTROL, 2007, Vo1.36, NO.3.
[4] Richard Yew Fatt Ng, Yong Hour Tay, Kai Ming Mok, "A
Review of Iris Recognition Algorithm", DOl: 978--4244-
2328-6/08.
[5] Ujwal\a Gawande, Mukesh Zaveri, Avichal Kapur,
Improving Iris Recognition Accuracy by Score Based
Fusion Method", International Journal of Advancements in
Technology (TJoAT) http://ijict.org/ ISSN 0976-4860.
[6] Daugman, 1., Dowing, C. "Epigenetic randomness,
complexity, and singularity of human iris patterns. In:
Preceding soft the Royal Society, B, 268, Biological
Sciences, pp.1737-1740(2001).
[7] Daugman, 1." Statistical richness of visual phase
information: update on recognizing Persons by iris patterns.
Int. 1. Comput.Vis.45 (1), 25-38(2001).
[8] John Daugman, "How Iris Recognition Works", IEEE
Transactions on Circuits and Systems for Video
Technology, Vol. 14, no. I, January 2004.
[9] Khattab M. Ali Alheeti, "Biometric Iris Recognition Based
on Hybrid Technique", International Journal on Soft
Computing ( IJSC ) Voi.2, No.4, November 2011 DOl :
10.512I!ijsc.2011.2401 1.