Attendance System Using Facial Recognition
Anuj Pant and
Srinivasan and Ms. P. Mahalakshmi
Computer Science and Engineering
Abstract—Face recognition is one of the most popular
applications of image understanding. The face recognition is a controversial
subject right now. A system such as this can recognize and track criminals
in a crowd, but some take it as an extreme invasion of privacy. Location Based Service (LBS) is critical to many businesses as well
as government organizations to fetch real time data from a location. The
spatial patterns that these services can provide is one of its most useful
aspect where location is a determining factor in these activities and can be
leveraged to better understand patterns and relationships. The prototype is
used to mobilize as well as increase the authenticity of the existing
attendance system by bringing the service directly to application environment.
We provide a deep analysis at the problem, draw conclusions and propose an
efficient executable solution to tackle these problems.
Index Terms—Facial Recognition, Location Based Services, Geolocation,
Database Management System, Application, Application Programming Interface.
A. Facial Recognition
A face recognition system is simply a computer
application capable of identifying or verifying a person from a digital image
or a video frame. Due to it’s nature, face recognition has recently gained
significant attention. Over the last decade or so, it has become a popular area
of research in computer vision and one of the most successful applications of
image analysis and understanding. One way to do this is by comparing selected
facial features from the face database. Face biometrics have the potential to
be integrated anywhere in a modern camera. Governmental organizations across
the world use biometric software to scan faces in CCTV footage, as well as to
identify persons of interest in the field. Defense agencies use face
recognition to verify the identities of tourists. Facial recognition doesn’t
just deal with hard identities, but also can gather data on public. This has
made face biometrics solutions much sought after in the retail marketing
industry. As a hands-free biometric solution that’s easy to deploy in consumer
devices, face recognition is showing the public just how convenient and simple
authentication can be.
B. Location Based Service
A location-based service
(LBS) is a software-level service that uses location data to operate certain
features in an application. As such LBS is an information service and has many
uses in social networking today as information, in entertainment or security,
which is accessible with mobile devices through the mobile network and which
uses information on the geographical position of the mobile device. Location
Based Service (LBS) is critical to many businesses as well as government
organizations to fetch real time data from a location. The spatial patterns
that location-related data and services can provide is one of its most useful
aspect where location is a determining factor in these activities and can be used
to better understand inter relationship between patterns. Geolocation is one of
the applications of LBS. It’s the estimation of the geographic location of an
object, such as a mobile phone, a street, or even a monument. Geolocation
involves the generation of a set of geographic coordinates, but its usefulness
is enhanced using these coordinates to determine a legit location, such as a
The application’s idea is general to all
the organizations using attendance system for the employees but for sake of the
demo, the university campus is taken as a sample workspace. Inside the campus,
when it’s time for classes to commence, we students generally observe the
faculty members rushing to biometrics system area for marking their attendance
and then going anywhere else. We
thought, that it would be so convenient that the faculty have this power within
their reach. This way they will save much time and energy, and efficient.
But the main issue was how to replace
biometrics with something that’s already incorporated in the mobile devices.
The clear answer to this was the front camera that today’s world user more than
often than the back camera for selfies. The front camera can be used for facial
recognition which is today’s one of the best tool for guaranteeing authenticity
of the user.
So, we thought of developing an app for
this task. The app would scan the user’s face and after successful facial
recognition, the user will be able to mark the attendance. But the catch here
is this, the user will be required to be present inside the campus premises.
Only then he/she will be able to mark his/her attendance thus ensuring their
physical presence inside the campus.
The project involves two technologies
known as Facial Recognition and Geolocation. With these two together with
Database Management System (DBMS), the project helps in marking attendance of
the teacher only when he/she is inside the campus. This way the authenticity of
the whole process is further improved on top of facial recognition.
The project aims at easing the lives of
the teaching staff by providing them with the facility to mark their attendance
right at their fingertips, without going to a area dedicated specifically for
biometrics recognition, thus saving them huge amount of time and energy.
This way, teachers could easily scan
their face, mark their attendance wherever they are within the campus premises.
They must create an account and upload their face initially, but it’s just a
one-time process for initial setup. The faculty can also check their attendance
anytime they want to keep track of it.
The attendance and all other credentials
of the user is saved at our 24-hour online hosted server, and is completely secure
and fast to access. Every action of the user inside the app is being kept track
of, so that there is no chance of fraud from user end.
Disclaimer: We are against stealing
user’s private information or any location details from their devices. We
strongly promote user privacy, and thus do not save any user’s information at
No matter how good the technology be,
there is always a chance of loopholes which needs to be covered gradually as
the technology is examined more thoroughly and checking every test case
possible. Same can be said for this project. These are some of the minor
loopholes that were found during the deployment of the project:
recognition technique used here is dependent on a 2D surface and not upon the
3D model. So, a user’s portrait or image on a flat surface when scanned, will
produce a successful result.
Sometimes it is possible to fake your current location to some other
location using some software hacks.
The above stated loopholes were
encountered during the application development. The problem stated above,
especially first, requires high amount of expertise in this field and huge
amount of developers to conquer it.
At this moment, the 3D facial recognition
is only used by Apple iPhone X in their device. Whilst, most of the other smartphone
giants still use 2D facial recognition which can be bypassed using a photo.
The application starts with the user entering in the details and his
initial face scan if they are a new user, or else they can straight away login
using their credentials they entered while creating their account. Once logged
in, the application checks the current location using the built in GPS system
of the mobile phone and then calculates the distance compared to the pre
entered co-ordinates of the campus. If the user is found to be within the
prescribed area or campus, he/she is eligible to mark the attendance, or else
the check-in button will be disabled until they get back into the campus area.
Once checked-in the user can view their attendance percentage till the
date all the way from their first day of joining. This parameter can be changed
to a monthly view or weekly, as per the requirement of the institution. The
attendance for the same is calculated using a virtual calendar created in our
database. This calendar can be modified as per the requirements of the
institution to feed in the working days and the days which may be marked as holidays.
Accordingly the attendance of each user is marked.
While leaving, the user has to simply press the check-out button in the
application and his out time will be marked while checking out, there is a
minimum number of working hours required, as in our case 8 hours. If the user
checks out early, he/she is asked for a valid reason, else the user may leave.
In some cases where in the user may work overtime or even forget to check-out
and may come back the next day to check-in and realise the same. In that case,
the attendance may or may not be given to the user as per the decision of the
institution and the case may be discussed with the Head of the Human Resources
in order to apply for overtime or mark as regular working day. As per the
decision the HR head may use their own interface to make the required changes
with respect to that candidate. The HR head may have access to the records of
all the users as well have the rights to make changes with respect to those
being partially marked due to longer entered working hours.
We Prakhar Saxena, and Anuj Pant, , would
like to express our gratitude to both out Faculty Advisors, Ms. P. Mahalakshmi and Mr. R.
Srinivasan as well as our Minor Project supervisor Mr. M. Senthil Raja for
the guidance throughout the Project without whose support this would not have
This project helped us to increase our
knowledge in the fields of Database handling as well as on how android
applications are developed. Also along the way we learnt an important lesson of the three D’s of
working Discipline, Determination and Devotion and help us throughout.
project was completed successfully with the help of the following references: