Journal of Computer Sciences and Applications, 2013, Vol. 1, No.

5, 100-105
Available online at http://pubs.sciepub.com/jcsa/1/5/4
© Science and Education Publishing
DOI:10.12691/jcsa-1-5-4

Fingerprint-Based Attendance Management System

Akinduyite C.O*, Adetunmbi A.O, Olabode O.O, Ibidunmoye E.O

Department of Computer Science,The Federal University of Technology, Akure, Ondo State, Nigeria
*Corresponding author: [email protected]

Received August 01, 2013; Revised August 12, 2013; Accepted November 04, 2013
Abstract In recent time, there has been high level of impersonation experienced on a daily basis in both private
and public sectors, the ghost worker syndrome which has become a menace across all tiers of government,
employers concerns over the levels of employee absence in their workforce and the difficulty in managing student
attendance during lecture periods. Fingerprints are a form of biometric identification which is unique and does not
change in one’s entire lifetime. This paper presents the attendance management system using fingerprint technology
in a university environment. It consists of two processes namely; enrolment and authentication. During enrolment,
the fingerprint of the user is captured and its unique features extracted and stored in a database along with the users
identity as a template for the subject. The unique features called minutiae points were extracted using the Crossing
Number (CN) method which extracts the ridge endings and bifurcations from the skeleton image by examining the
local neighborhoods of each ridge pixel using a 3 x 3 window. During authentication, the fingerprint of the user is
captured again and the extracted features compared with the template in the database to determine a match before
attendance is made. The fingerprint-based attendance management system was implemented with Microsoft’s C# on
the. NET framework and Microsoft’s Structured Query Language (SQL) Server 2005 as the backend. The
experimental result shows that the developed system is highly efficient in the verification of users fingerprint with an
accuracy level of 97.4%. The average execution time for the developed system was 4.29 seconds as against 18.48
seconds for the existing system. Moreover, the result shows a well secured and reliable system capable of preventing
impersonation.
Keywords: fingerprint, attendance management, enrolment, authentication, Crossing Number, minutiae score
Cite This Article: Akinduyite C.O, Adetunmbi A.O, Olabode O.O, and Ibidunmoye E.O, “Fingerprint-Based
Attendance Management System.” Journal of Computer Sciences and Applications 1, no. 5 (2013): 100-105. doi:
10.12691/jcsa-1-5-4.

automatic personal identification. The reason for the

popularity of fingerprint verification is that fingerprints
1. Introduction satisfy uniqueness, stability, permanency and easily taking [3].
In this paper, an attempt was made to look at the
It is expected today that an individual who wants to prevalence in the high level of impersonation experienced
authenticate himself for a service must have a token on a daily basis in both private and public sectors, the
and/or password for example identity card, ATM card, ghost worker syndrome which has become a menace
driving license, health card and so on. Carrying different across all tiers of government, employers concerns over
cards and remembering passwords for different services is the levels of absence in their workforce and difficulty in
a significant issue for individuals and organizations. A managing student attendance during lecture periods.
secure and effective identity management system plays an Sequel to this, a fingerprint-based Attendance
important role in the successful deployment of an attendance Management System was developed to provide a faster,
management system. To make the identity management more secure, and more convenient method of user
system more secure and reliable for authentication, biometrics verification than passwords and tokens can provide for a
data are integrated in the attendance management systems [1]. reliable personal identification.
Biometrics technologies verify identity through
characteristics such as fingerprints, faces, irises, retinal
patterns, palm prints, voice, hand-written signatures, and 2. Attendance Management
so on. These techniques, which use physical data, are
receiving attention as a personal authentication method Attendance management is the act of managing
that is more convenient than conventional methods such as attendance or presence in a work setting to minimize loss
a password or ID cards because it uses data taken from due to employee downtime. Attendance control has
measurements and such data is unique to the individual traditionally been approached using time clocks and
and remains so throughout one’s lifetime [2]. timesheets, but attendance management goes beyond this
In these technologies, fingerprint becomes the most to provide a working environment which maximizes and
mature and popular biometrics technology used in motivates employee attendance [4].
 Journal of Computer Sciences and Applications 101

Attendance management is a major part of today’s software that is paired with a time clock for employees
human resource systems; take organization towards better which uses biometric technology for authentication
human resource practice, systems and excellence, hence purposes. When these systems are in use, the employees
regular attendance and punctuality are expected of all can use their biometric data such as finger prints for
employees or candidates in a work setting. Unsatisfactory clocking in and clocking out. This method has the great
attendance caused by unscheduled absences and tardiness benefit that the entire process is easy as well as quick.
cause a disruption in work, affects productivity, and Other advantages include elimination of the cost
creates morale problems when workloads are shifted to previously incurred in getting the employees cards.
other employees [5].
Moreover, in many institutions, and academic
organizations, attendance is also a very important criteria 3. Related Works
which is used for various purposes. These purposes
include record keeping, assessment of students, and [7] proposed an embedded computer-based lecture
promotion of optimal and consistent attendance in class. attendance management system where a single-chip
In developing countries, a minimum percentage of class computer based subsystems (an improvised electronic card
attendance is required in most institutions and this policy and the card reader) were interfaced serially to the serial
has not been adhered to, because of the various challenges port of the digital computer. The electronic card is a
the present method of taking attendance presents. This model of a smart card containing the student identity (ID-
traditional method involves the use of sheets of paper or Name, Matriculation Number and five pin encrypted code).
books in taking student attendance. This method could The student ID is authenticated by the card reader which
easily allow for impersonation and the attendance sheet compares the entrance code with the encrypted code on
could be stolen or lost. Taking of attendance is time the card swiped through the card reader. The student is
consuming and it is difficult to ascertain the number of granted and/or denies specific lecture attendance based on
students that have made the minimum percentage and thus the result of the comparison by the backend software
eligible for exam. Thus, there is a need for a system that system running on the PC to which the card reader is
would eliminate all of these trouble spots. serially interfaced. The system though provided a
simplified, low cost embedded computer based system
2.1. Types of Attendance Management solution to the management of lecture attendance problem
in developing countries but does not eliminate the risk of
System impersonation. The system is devise-based in which
Attendance Management falls into two categories students have to carry RFID cards and also the RFID
namely; Conventional and Automated methods. detectors are needed to be installed.
Conventional methods include time sheet, attendance [8] proposed a real time computer vision algorithms in
register and time clock. Time sheets are documents, automatic attendance management systems using
electronic or otherwise that record what time was spent by Computer vision and face recognition algorithms and
the employee on what tasks. Attendance register is an integrating both into the process of attendance
official list of people who are present at an institution or management. The system eliminates classical student
organisation. Time clock which is a mechanical (or identification such as calling student names, or checking
electronic) time piece used to assist in tracking the hour respective identification cards, but still lacks the ability to
worked by an employee of a company. identify each student present in class thereby providing a
Automated methods include Barcode system attendance lower recognition rate because facial images are subject to
system, magnetic stripe attendance system, Radio change between the time of enrolment and time of
Frequency Identification (RFID) and the biometric verification and also poses a bigger financial burden during
attendance system [6]. installation and does not offer any privacy protection.
The barcode attendance system requires that every In [2], a wireless attendance management system based
employee is issued a badge/card in which there is a on iris recognition was proposed using Daugman’s
barcode. In order to check into or out of the company, the algorithm. The system uses an off-line iris recognition
badge/card is swapped on the time clock, and the data is management system that can finish all the process
captured by the clock. In the magnetic stripe attendance including capturing the image of iris recognition,
system, data is encoded in the magnetic stripe of the extracting minutiae, storing and matching but it is difficult
employee card. When the card, is swiped through the to lay the transmission lines where topography is bad.
employee time clock, the information in the card’s
magnetic stripe is recorded by the time clock. This system
reads one card at a time and also requires contact with the 4. System Overview
reader. Radio-frequency identification (RFID) is a
technology that uses radio waves to transfer data from an This proposed system introduces a new automatic
electronic tag, called RFID tag or label, attached to an attendance management system, which integrates
object, through a reader for the purpose of identifying and fingerprint authentication into the process of attendance
tracking the object. The ID cards of the employees is management for both staff and student. It is made up of
embedded with RFID tag which is read by a reader. This two processes namely; enrolment and authentication.
RFID system is interfaced to a database through a During enrolment, the biometrics of the user is captured
computer. Each employee uses an RFID card and the and the minutiae data are extracted and stored in a
reader records the data when the employee enters or exits. database as a template for the subject along with the user’s
In biometric Attendance system, there is attendance ID. The objective of the enrolment module is to admit a
102 Journal of Computer Sciences and Applications

user using his/her ID and fingerprints into a database after determine the identity of the user, formulating the process
feature extraction. These features form a template that is of authentication. The enrolment process is carried out by
used to determine the identity of the user, formulating the an administrator of the attendance management system.
process of authentication. The enrolment process is carried The enrolment and registration phase is an administrative
out by an administrator of the attendance management phase. The user fingerprint as well as other bio-data is
system. During authentication, the biometrics of the user stored for the first time into the database for registration.
is captured again and the extracted features are compared The courses, practicals, tests, lecturers and exams are also
with the ones already existing in the database to determine registered at this phase. All data and information required
a match. After a successful match, attendance is marked for the proper recording of attendance are enrolled in this
against the user’s id used in matching the templates. module.
The work utilized a fingerprint reader as the input to The most commonly employed method of minutiae
acquire images, developed program that has fingerprint extraction is the Crossing Number (CN) [9] upon which
recognition and identification system as well as database this research is based. It involves the use of the skeleton
to store user’s information. The database comprises the image where the ridge flow pattern is eight-connected.
fingerprint templates and other bio-data of the users The minutiae are extracted by scanning the local
together with the attendance records made by the users. neighbourhood of each ridge pixel in the image using a 3 x
Figure 1 shows the architecture of the proposed attendance 3 window. The CN value is then computed, which is
management system. defined as half the sum of the differences between pairs of
adjacent pixels in the eight-connectivity neighbourhood.
The CN for a ridge pixel P is given by
8
CN =0.5∑ Pi − Pi +1 P9 =P1 (1)
i −1

where Pi is the pixel value in the neighbourhood of P.

5.2. Authentication Module

The task of the authentication module is to validate the
identity of the person who intends to access the system.
The person to be authenticated indicates his/her identity
Figure 1. Architecture of the proposed fingerprint-based attendance and places his/her finger on the fingerprint scanner. The
management system fingerprint images captured is enhanced and thinned at the
image processing stage, and at feature extraction stage, the
biometric template is extracted. It is then fed to a
5. System Architecture matching algorithm, which matches it against the person’s
biometric template stored in the system database to
The design of the fingerprint-based attendance establish the identity. During authentication, for staff
management system is made up of the following: attendance, a staff supply his/her department and name,
ⅰ. Enrolment module then places his/her finger over the fingerprint reader, the
ⅱ. Authentication Module fingerprint recognition unit compares the fingerprint
ⅲ. System database. features with those stored in the database, after a
successful match, the staff’s employee number is sent to
5.1. Enrolment Module the database alongside the time of making such an
attendance and update the status (either present/absent) of
The task of enrollment module is to enroll users and
user’s attendance for the day. Staff attendance is captured
their fingerprints into the system database. During
twice a day for both arrival and departure time.
enrolment, the fingerprint and other bio-data of the user is
For student attendance, the lecturer (or a designated
captured and the unique features are extracted from the
personnel as the case may be) selects his/her department,
fingerprint image and stored in a database as a template
level, course code, attendance type (for example lecture,
for the subject along with the user’s ID. Staff bio data to
practicals etc) and the attendance ID, then the student
be captured includes: employee number, surname, other
places his/her fingerprint on the fingerprint reader; the
names, sex, position, staff type, phone number, email,
fingerprint recognition unit compares the fingerprint
department and passport photograph. Student bio data
features with those stored in the database, after a
includes: matriculation number, surname, other-names,
successful match, the student’s matriculation number is
sex, department, level, studentship, phone number and
sent to the database alongside the time of making such
passport photograph. To improve the quality of a captured
attendance and update the status (either present/absent) of
image during enrolment/registration, two image samples
student’s attendance for the class. Student attendance is
per fingerprint used are captured for a higher degree of
captured only once for each attendance type.
accuracy.
Fingerprint matching approaches includes minutiae-
When the fingerprint images and the user name of a
based matching, ridge-based matching and the correlation
person to be enrolled are fed to the enrollment module, a
matching approaches. However, it is believed that
minutiae extraction algorithm is first applied to the
minutiae-based matching approach, upon which this work
fingerprint images and the minutiae patterns (features) are
is based facilitates the design of a robust, simple, and fast
extracted. These features form a template that is used to
verification algorithm while maintaining a small template
 Journal of Computer Sciences and Applications 103

T T
size. Minutiae-based representation is commonly used, rowref , colref = row index and column index of reference
primarily because forensic examiners have successfully T T
relied on minutiae to match fingerprints for more than a points currently being considered. row and col
k k
century, minutiae-based representation is storage efficient,
represents row index and the column index of the kth
and expert testimony about suspect identity based on
minutiae. Similarly the input matrix data points are
mated minutiae is admissible in courts of law [10].
converted to polar coordinates using the Equation 5.
Most common minutiae matching algorithms consider
each minutiae as a triplet m = {x, y, θ} that indicates the  
( ) ( )
2 2
 rowm − rowref + colm − colref
I I I I
(x, y) minutiae location coordinates and the minutiae 
angle θ [11]. Extracted minutiae from the fingerprint are  γ mI   
together forming a point pattern in plane. Therefore     rowmI − rowref
I  
=  φ I   tan −1   + rotatevalues (k , m)  (5)
matching two minutiae point patterns with each other are  m   colmI − colref
T  
considered as a 2D point pattern problem. The point  θ mI     
   T 
patterns are constructed only on positions (x, y) of
 θ k − θ ref
T

minutiae in the plane. Since point patterns are based on
 
positions of minutiae in fingerprint they form distinctive
patterns. With enough points in each pattern the positions Where for an input image,
I
(x, y) of the minutiae are the only information that is rm = radial distance of mth minutiae.
needed for good matching results. I
Let T and I be the representation of the template and ϕ m = radial angle of mth minutiae.
I
input fingerprint, respectively. Let the minutiae sets of the θm = orientation angle of mth minutiae.
template be given as: I I
rowref , colref = row index and column index of reference
= T {m1 , m2 , = mm } mi {= xi , yi , ϑi } i 1 m (2) points currently being considered. Rotate values (k, m)
and the minutiae sets of the input fingerprint be given as : represents the difference between the orientation angles of
I I

{=
x j y jϑ j } j
Tk and Im. row and col represents row index and the
=l {m1 , m2 
= mm } m j 1 m (3) m m
column index of the mth minutiae. Tk and Im represent the
A minutia mj in I and a minutia mi in T are considered extracted data in all the columns of row k and row m in the
to be matched if their spatial and orientation differences template and input matrices, respectively.
are within specified thresholds ro and θo. For efficient
matching process, the extracted data is stored in the matrix 5.3. The Database
format [12] as follows.
The attendance management system database consists
Number of rows: Number of minutiae points.
of tables that stores records, each of which corresponds to
Number of columns: 4
an authorized person that has access to the system. Each
Column 1: Row index of each minutiae point.
record may contain the minutiae templates of the person’s
Column 2: Column index of each minutiae point.
fingerprint and user name of the person or other
Column 3: Orientation angle of each minutiae point.
information such as pin no as an index to the template.
Column 4: Type of minutiae. (A value of ‘1’ is assigned
The database design for the system implements relational
for termination, and ‘3’ is assigned for bifurcation).
data model which is a collections of tables in which data
During the matching process, each input minutiae point
are stored. The database was implemented in Microsoft
is compared with template minutiae point. In each case,
SQLServer database (Sql Server, 2005). SQLServer is fast
template and input minutiae are selected as reference
and easy, it can store a very large record and requires little
points for their respective data sets. The reference points
configuration.
are used to convert the remaining data points to polar
coordinates. The Equation 4 is used to convert the
template minutiae from row and column indices to polar
coordinates.
6. System Performance and Evaluation
  Given a fingerprint matcher, one would like to assess its
( ) ( colkT − colrefT )
T 2 2
 rowk − rowref +
T
 accuracy and speed performance in a realistic setting.
 γ kT    Unlike passwords and cryptographic keys, biometric
    rowkT − rowref
T   templates have high uncertainty. There is considerable
 φkT  =  tan −1    (4) variation between biometric samples of the same user
     
 colk − colref
T T
 θ kT     taken at different instances of time. Therefore the match is
   T  always done probabilistically. This is in contrast to exact
 θ k − θ ref
T
 match required by password and token based approaches.
  The inexact matching leads to two forms of errors namely:
Where for a template image, False (impostor) Acceptance Rate (FAR) and the False
T
rk = radial distance of kth minutiae. (genuine individual) Rejection Rate(FRR). The FAR/FRR
I ratios depend, among other factors, on the type of
ϕ m = radial angle of kth minutiae. difficulty of the algorithms used in the fingerprint
T extraction. Usually, algorithms with high-medium
θk = orientation angle of kth minutiae.
complexity lead to acceptable low FRR/FAR [13].
104 Journal of Computer Sciences and Applications

False Accept: An impostor may sometime be accepted matched against the already stored template in the
as a genuine user, if the similarity with his template falls database. In each case, if the matching score is less than
within the intra-user variation of the genuine user. The the threshold, the person is rejected otherwise the person
FAR normally states, either in a percentage or a fraction, is accepted. Using equation 6-8, Table 1 gives the
the probability of someone else matching as you. FAR is respective values for the false Acceptance rate (FAR) and
defined by the formula: False Rejection Rate (FRR) and for the test that was
carried out.
FA
=
FAR ∗100 (6) In the test, the false acceptance rate was zero meaning
N that there were no cases of false acceptance (FAR) i.e. a
Where FA is the number of false accept and N is the total person that was not pre-registered was not falsely enrolled
number of verification. for attendance. There were a few false rejections (FRR)
False Reject: When the acquired biometric signal is of during the test in which the system failed to identify some
poor quality, even a genuine user may be rejected during pre-registered users. These could be attributed to improper
authentication. This form of error is labelled as a ’false placement of the finger on the scanner and fingers that
reject’. If you fail to match against your own template, have been slightly scarred due to injuries.
then you have been falsely rejected. The probability of this
Table 1. Values of the FAR and FRR from the Test
happening is referred to as the false rejection rate, or FRR. FAR FRR
Thus, the higher the probability of false rejection, the 0.0% 2.56%
greater the likelihood you will be rejected. FRR is defined
The above values of FAR and FRR implies an accuracy
by the formula:
of 97.4% considering the genuine acceptance rate. Tables
FR 2a and 2b show the details.
=
FRR ∗100
N (7) Table 2a. Details of the evaluation
Where FR is the number of false reject and N is the Successful Unsuccessful
Verification Verification
total number of verification.
Staff 30 0
The system may also have other less frequent form of Student 84 3
error such as: Total 114 3
• Failure to enroll(FTE) It is estimated that nearly 4%
of the population have illegible fingerprints. This Table 2b. Details of the evaluation
consists of senior population, laborers who use their No of Successful Unsuccessful
Accuracy
Attendee Verification Verification
hands a lot and injured individuals. Due to the poor
ridge structure present in such individuals, such users 117 114 3 97.4%
cannot be enrolled into the database and therefore
cannot be subsequently authenticated. The FTE Table 3. Time taken for Verification
normally states, either in a percentage or a fraction, Total Total
Average
the possibility of someone failing to enroll in a No of Execution
Type of system time in time in
attendee Time in
system. seconds minutes
seconds
Fingerprint-Based
FE
(8)
117 502.41 8.37 4.29
=
FTE ∗100 Attendance System
N Manual Attendance
117 2161.55 36.02 18.48
System
Where FE is the total number of Failure Enrol and N is the
total number of verification.
For performance analysis, the application developed
was tested using the bio-data and fingerprints collected
from One hundred and seventeen (117) users out of which
30 were staff and 87 were students of the department of
computer science, the Federal University of Technology,
Akure, Nigeria using a live-scan method. The fingerprints
were taken from any of the ten fingers of a respective
member of the group in which each person must
remember the exact finger that was used for the purpose of
verification. The reason for this is to allow alternate finger
should any of the fingers fails to enrol due to the poor
ridge structure present in such fingers. The reason for
interest in the use of students and staff of the above
mentioned department and school respectively is easy
accessibility and their readiness to provide their biometric Figure 2. Comparison of the Manual Attendance System with the
Fingerprint-based Attendance Management System.
data for research purposes.
The minutiae data were extracted from the fingerprint The developed fingerprint-based attendance management
images and stored in a database as a template for the system was compared with the existing manual attendance
subject along with the user’s ID. During authentication, system (use of paper sheet/attendance register) and the
the biometric of the user is captured again and minutiae time of taking both attendance was recorded. The manual
data are also extracted forming the test template which is attendance system average execution time for one hundred
 Journal of Computer Sciences and Applications 105

and seventeen (117) attendees is approximately 18.48 References

seconds as against 4.29 seconds for the fingerprint-based
attendance management. Table 3 shows the time taken for [1] EPIC-Electronic Privacy Information Centre (2002):“National ID
verification. Figure 2 shows the comparison of the manual Cards,” http://www.epic.org/privacy/id_cards/. accessed January,
system with the Fingerprint-based Attendance Management 2012.
System. [2] Kadry S. and Smaili M. (2010): Wireless Attendance Management
System based on Iris Recognition. Scientific Research and Essays
Vol. 5(12), pp. 1428-1435, 18 June, 2010.
[3] Khan B., Khan M. K. and Alghathbar K. S. (2010): Biometrics
7. Conclusion and identity management for homeland security applications in
Saudi Arabia. African Journal of Business Management Vol. 4(15),
In this paper, we have presented a fingerprint-based pp. 3296-3306, 4 November, 2010.
[4] Bevan S and Hayday S. (1998): Attendance Management: a
attendance management system. The developed system is Review of Good Practice" Report 353, Institute for Employment
an embedded system that is part of a fingerprint Studies.
recognition/authentication system based on minutiae [5] McKeehan D.A. (2002): Attendance Management Program, The
points. The system extract the local characteristic of a City of Pleasanton, Human Resources.
fingerprint which is minutiae points in template based. [6] Ononiwu G. C and Okorafor G. N (2012): Radio Frequency
Identification (RFID) Based Attendance System With Automatic
Templates are matched during both registration and Door Unit, Academic Research International. Vol 2, No 2, March,
verification processes. For improved quality control 2012.
during the registration or verification process, a matching [7] Shoewu O., Olaniyi O.M. and Lawson A. (2011): Embedded
score was used to determine the success of the operation. Computer-Based Lecture Attendance Management System.
African Journal of Computing and ICT. Vol 4, No. 3. P 27- 36,
The matching score was specified so that only sets of
September, 2011.
minutiae data that exceed the score will be accepted and [8] Shehu V. and Dika A. (2011): Using Real Time Computer Vision
data below the score will be rejected. Therefore, Algorithms in Automatic Attendance Management Systems.
Fingerprint Recognition using Minutia Score Matching Proceedings of the ITI 2010 32nd Int. Conf. on Information
method was used for matching the minutia points before Technology Interfaces, June 21-24, 2010, Cavtat, Croatia.
attendance is recorded. [9] Mehtre, B. M. (1993): Fingerprint image analysis for automatic
identification. Machine Vision and Applications 6, 2 (1993), 124-
The developed system is very helpful in saving valuable 139.
time of students and lecturers, paper and generating report [10] Jain A. K., Maio D., Maltoni D., and Prabhakar S. (2003):
at required time. The system can record the clock in and Handbook of Fingerprint Recognition, Springer, New York, 2003.
clock out time of students and workers in a very [11] Maltoni D. and Cappelli R. (2008): Fingerprint Recognition, In
Handbook of Biometrics, Springer Science + Business Media,
convenient manner using their fingerprint to prevent U.S.A.
impersonation and reduce level of absence. Also, it [12] Ravi. J. K., Raja b. and Venugopal. K. R.(2009): Fingerprint
reduces most of the administrative jobs and minimizes Recognition Using Minutia Score Matching, International Journal
human errors, avoids proxy punching, eliminates time- of Engineering Science and Technology Vol.1(2), 2009, 35-42.
related disputes and helps to update and maintain [13] Sharat S. Chikkerur(2005): Online Fingerprint Verification
System, A MTech Thesis, Department of Electrical Engineering,
attendance records. Faculty of the Graduate School of the State University of New
York at Buffalo.