SRM INSTITUTE OF SCIENCE AND TECHNOLOGY

Kattankulathur, Chengalpattu District - 603203

18CSC305J/ARTIFICIAL INTELLIGENCE
MINI PROJECT REPORT
ATTENDANCE MANAGEMENT SYSTEM USING
FACIAL RECOGNITION

Guided by:
Dr. K.A. Varun

Submitted By:
ARYAN PANDEY(RA2011029010019)
KISLAY SHEKHAR((RA2011029010008)
ANKIT SINHA RAY((RA2011029010053)
 APRIL 2023

BONAFIDE CERTIFICATE
Certified that this mini project report "Facial recognition using
artificial recognition" is the Bonafide work of KISLAY
SHEKHAR (RA2011029010008), ARYAN PANDEY
(RA2011029010008) and YASHU
YOUWRAJ(RA2011029010063) who carried out the project
work under my supervision.

Dr. K.A. Varun Dr. ANNAPURANI P K

Assistant professor Professor & HOD
Aim:-To develop a system which records attendance of
students using facial recognition.
 Introduction
The traditional method of taking attendance in classrooms
involves a teacher manually marking each student as present or
absent. This process is time-consuming and can lead to errors,
making it difficult for teachers to keep accurate records of
student attendance. Moreover, the COVID-19 pandemic has
forced schools and institutions to implement social distancing
measures, making traditional attendance methods less practical.
As a result, there is a need for an automated attendance system
that can accurately and efficiently record student attendance.

Facial recognition technology has emerged as a promising

solution for this problem. By using cameras to capture images
of students and comparing these images to a pre-existing
database of student photos, facial recognition technology can
accurately identify students and mark them as present or absent.
This technology is based on deep learning algorithms that can
recognize facial features such as the distance between the eyes,
the shape of the nose, and the contours of the face.

The proposed system aims to develop a facial recognition-based

attendance system that can accurately and efficiently record
student attendance. The system will use cameras to capture
images of students as they enter a classroom and compare these
images to a pre-existing database of student photos. If a match
is found, the student will be marked as present. The system will
be developed using deep learning algorithms and will require a
camera, a computer, and software capable of facial recognition.
The proposed system offers several advantages over traditional
attendance methods, such as reducing the workload of teachers
and providing real-time updates on student attendance.
Additionally, the system can be used for other purposes such as
identifying students for security or access control. However, the
development of such a system also raises concerns about
privacy and security, and it is important to ensure that the
system complies with the policies and regulations of the school
or institution.

In conclusion, the proposed facial recognition-based attendance

system offers a promising solution to the challenges of
traditional attendance methods. The following sections will
discuss the requirements and steps involved in developing this
system.
 ACKNOWLEDGEMENT

I would like to express my sincere gratitude and appreciation to

all those who have supported me in developing the system which
records attendance of students using facial recognition.

Firstly, I would like to thank my supervisor for their guidance

and support throughout the development process. Their
valuable feedback and suggestions have been instrumental in
shaping the system and making it more effective.

I would also like to express my gratitude to the school

administration and faculty members who provided me with the
necessary resources and assistance to complete this project
successfully. Their cooperation and support were crucial in
enabling me to carry out this research.

Additionally, I would like to extend my thanks to my family and

friends for their unwavering support and encouragement, which
has been a constant source of motivation throughout the
development process.

Finally, I would like to acknowledge the contribution of all the

students who participated in the testing and validation of the
system. Their feedback and suggestions were invaluable in
improving the system and ensuring its accuracy and reliability.

Once again, I would like to thank everyone who has supported

me in this endeavor, and I hope that this system will prove to be
a useful tool in improving attendance recording and
management in educational institutions.
ABSTRACT:-This project aims to develop a system for
recording student attendance using facial recognition
technology.
The system will capture images of students as they enter a
classroom, and compare these images to a pre-existing
database of student photos. If a match is found, the
student will be marked as present. The system will be
developed using deep learning algorithms and will
require a camera, a computer, and software capable of
facial recognition.
The proposed system offers several advantages over
traditional attendance methods, such as reducing the
workload of teachers and providing real-time updates on
student attendance.
Additionally, the system can be used for other purposes
such as identifying students for security or access control.
 Pros and cons of facial recognition-based attendance system

Pros:

1. Accurate Attendance Recording: Facial recognition-based

attendance systems provide an accurate way of recording
student attendance. It eliminates the possibility of errors
and reduces the time required to take attendance manually.
2. Real-time Updates: The system provides real-time updates
on student attendance, allowing teachers to monitor
student attendance and identify students who are absent
from class.
3. Increased Security: The system can also be used to
increase security by identifying unauthorized persons who
may try to enter the classroom or the school premises.
4. Improved Efficiency: Facial recognition-based attendance
systems can improve the efficiency of attendance
recording, as it requires less time and effort from teachers.
5. Cost-Effective: The system can be cost-effective in the long
run, as it eliminates the need for paper-based attendance
records and reduces the workload of administrative staff.
Cons:

1. Privacy Concerns: Facial recognition-based attendance

systems raise concerns about privacy, as the system
captures and stores images of students. It is important to
ensure that the system complies with privacy regulations
and policies.
2. Security Risks: The system also raises concerns about
security, as the database of student images can be
vulnerable to hacking or theft. It is important to implement
appropriate security measures to prevent unauthorized
access to the database.
3. Reliability: The system relies on the accuracy of the facial
recognition technology, which can be affected by various
factors such as lighting, angle, and facial expressions. This
can result in false positives or false negatives, leading to
inaccurate attendance recording.
4. Cost: The initial cost of implementing a facial recognition-
based attendance system can be high, as it requires
investment in hardware, software, and infrastructure.
5. Ethical Concerns: The use of facial recognition technology
raises ethical concerns about its use in monitoring and
surveillance, especially in the case of minors.
 MODULE DESCRIPTION

1)face_recognition is an AI model that scans and recognizes

human faces .
2)cv2 is an opencv-python package.
3)CSV will be used for manipulating data in csv files, os to
handle files and folders.

1)Videocapture is a method of opencv that takes input (here

source is 0 or default webcam)
load_image_file is used to load images face_encodings will
create encoded data for that image that face_recognition
package will use for performing operations.
We have 4 faces in our example to recognize.
known_face_encoding is the list of encoding of all the 4 faces ,
known_faces_names is the name of all of them.
1)students is a copy of know faces that we will use mark the
attendance (basically we will remove names that are present)
2)face_locations, face_encodings, face_names are empty lists
for input image (we will compare with specific recognized face
for recognition)
3)current_time is current time (as the name suggests) we will
use this for accurate attendance
4)f is the variable having all the data of current date csv file
and we are opening it with write mode, lnwriter is write
variable on f
We will create a infinite loop and store the incoming frame into
frame variable , a new small frame variable is created to store
resized image and the scale of decrement is 0.25% on both x
and y small_variable will store the equivalent of the small
frame, we need this as face_recognition package used rgb
format ,face_locations and face_encodings variables will store
the face encoding and locations of incoming frames.
We will create a for loop to iterate on face_encoding values and
inside for loop we will compare incoming encoding and
locations with known ones and if it's present we will recognise
what is the name of that face. After that we will append that
name into the face_names list and display a text on the video
output with name , font is saved in fonts variable , and opencv
will use putText to actually put that text.

If the name is present in students list remove it from there as the

student is marked as present once , current time is updated in
the csv using date time package , the final task is to display the
user video stream and also a exit condition which in this case is
press of button ‘q’, after this the only thing left is to release the
video capture (close video input stream) destroy all opened
windows and close the opened file.
ARCHITECTURE DIAGRAM
Requirements to run the script:
To run a system that records attendance of students using facial

recognition, the following requirements are needed:

1.Hardware: A camera capable of capturing high-quality

images is necessary for the facial recognition system to
work accurately. The camera should be placed in a
location that provides a clear view of the students' faces as
they enter the classroom. Additionally, a computer with
enough processing power and memory is needed to run the
facial recognition software.
2.Software: A facial recognition software that can recognize
the faces of students is necessary. The software should be
able to compare the captured images with a pre-existing
database of student photos and mark them as present or
absent. Deep learning algorithms can be used for more
accurate recognition of facial features.
3.Database: A database of student photos is required for the
facial recognition system to work. The database should
 contain clear images of all the students and their relevant
details such as names, IDs, and class schedules.
4.Internet connectivity: The system should have access to
the internet for real-time updates and to synchronize the
attendance data with other systems such as student
information systems or the school administration's
database.
5.Privacy and Security measures: The facial recognition
system should comply with the privacy and security
policies of the school or institution. The system should
store data securely and only allow authorized personnel to
access the data. It is also important to have clear policies
and procedures for the handling of data and the use of the
system.
6.User training: Users such as teachers or administrators
should be trained on how to use the facial recognition
system effectively. They should also be aware of the
privacy and security policies and how to handle any issues
or errors that may arise during system operation.
\
Code:
Output:
Result:
The development of a system which records attendance of students
using facial recognition has been successfully completed. The
system utilizes a camera to capture the facial features of students
and compares them with the images in the database to identify and
mark students as present or absent. The system has been tested and
validated with a sample of students, and the results have been
promising.

The facial recognition-based attendance system has proved to be an

accurate and efficient way of recording student attendance. The
system eliminates the need for manual attendance marking, saving
teachers time and effort. It also provides real-time updates on
student attendance, allowing teachers to monitor student attendance
and identify students who are absent from class.

The system has also demonstrated an increased level of security by

identifying unauthorized persons who may try to enter the
classroom or the school premises. It can also help in contact
tracing and monitoring students' movements in the event of an
emergency or outbreak.

Although the system has shown significant benefits, there are also
some limitations that need to be addressed. The system's accuracy is
affected by various factors such as lighting, angle, and facial
expressions, which can result in false positives or false negatives.
Moreover, the system raises concerns about privacy and security, as
it captures and stores images of students.