A PROJECT REPORT ON

AI BASED ATTENDANCE SYSTEM USING FACE

RECOGNITION

SUBMITTED TO MIT SCHOOL OF ENGINEERING

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF
THE DEGREE

BACHELOR OF TECHNOLOGY
(Computer Science & Engineering)

BY

Shreyash P. Kalbhor MITU19BTCS0135

Siddharth S Desai MITU19BTCS0080
Saurabh Vidhyarthi MITU19BTCS0299
Rishikesh N Gaikwad MITU19BTCS0095

Under The Guidance of

Prof. Sachin Jagadale

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

MIT School of Engineering
MIT Art, Design and Technology University,
Rajbaug Campus, Loni-Kalbhor, Pune 412201
2022-23
 MIT SCHOOL OF ENGINEERING
DEPARTMENT OF COMPUTER SCIENCE AND
ENGINEERING MIT ART, DESIGN AND TECHNOLOGY
UNIVERSITY, RAJBAUG CAMPUS, LONI-KALBHOR, PUNE
412201

CERTIFICATE

This is to certify that the Project Entitled: -

AI BASED ATTENDANCE SYSTEM USING FACE

RECOGNITION
Submitted by

Shreyash P. Kalbhor MITU19BTCS0135

Siddharth. S. Desai MITU19BTCS0080
Saurabh Vidhyarthi MITU19BTCS0299
Rishikesh. N. Gaikwad MITU19BTCS0095

is a bonafide work carried out by them under the supervision of Prof. Sachin
Jagadale and it is submitted towards the partial fulfillment of the requirement of
MIT ADT University, Pune for the award of the degree of Bachelor of Technology
(Computer Science and Engineering).

Prof. Sachin Jagadale Dr. Ganesh Pathak Dr. Rajneeshkaur Sachdeo

Internal Guide H.O.D Director
Department of CSE Department of CSE MIT SoE
 CERTIFICATE

This is to certify that the Project Entitled

AI BASED ATTENDANCE SYSTEM USING FACE

RECOGNITION
Submitted by

Shreyash P. Kalbhor MITU19BTCS0135

Siddharth. S. Desai MITU19BTCS0080
Saurabh Vidhyarthi MITU19BTCS0299
Rishikesh. N. Gaikwad MITU19BTCS0095

is a bonafide work carried out by them under the supervision of Prof. Sachin Jagadale and has

been completed successfully.

Mr/Mrs. _________________________
Designation: _____________________
External Guide: ___________________

Seal and Stamp of the company

 DECLARATION

We, the team members

Shreyash P. Kalbhor MITU19BTCS0135

Siddharth. S. Desai MITU19BTCS0080
Saurabh Vidhyarthi MITU19BTCS0299
Rishikesh. N. Gaikwad MITU19BTCS0095

Hereby declare that the project work incorporated in the present project entitled
AI Based Attendance System Using Face Recognition is original work. This work (in
part or in full) has not been submitted to any University for the award or a Degree
or Diploma. We have properly acknowledged the material collected from
secondary sources wherever required. We solely own the responsibility for the
originality of the entire content.

Date: /05/2023

Name & Signature of the Team Members

Shreyash P Kalbhor _________________

Siddharth S Desai _________________

Saurabh Vidhyarthi _________________

Rishikesh N Gaikwad _________________

Name and Signature of Guide

Prof. Sachin Jagadale _________________

Seal/Stamp of the college

Place: MIT ADT University
Date: / 05 / 2023
 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
MIT SCHOOL OF ENGINEERING
RAJBAUG, LONI KALBHOR
PUNE – 412201

EXAMINER’S APPROVAL CERTIFICATE

The project report entitled AI Based Attendance System Using Face Recognition
submitted by Shreyash P Kalbhor (MITU19BTCS0135), Siddharth S
Desai((MITU19BTCS0080), Saurabh Vidyarthi (MITU19BTCS0299), Rishikesh N Gaikwad
(MITU19BTCS0095) in partial fulfilment for the award of the degree of Bachelor of
Technology (Computer Science & Engineering) during the academic year 2022-23, of
MIT-ADT University, MIT School of Engineering, Pune, is hereby approved.

Examiners

Examiner 1: ________________

Examiner 2: ________________
 Acknowledgments

It gives us great pleasure in presenting the project report on ‘AI Based Attendance System
Using Face Recognition’.

We would like to take this opportunity to thank my internal guide, Prof. Sachin Jagadale,
for giving me all the help and guidance I needed. I am really grateful to them for their kind
support. Their valuable suggestions were very helpful.

We are also grateful to Dr. Ganesh Pathak, Head of Computer Science & Engineering
indispensable support, suggestions.

Shreyash P Kalbhor
Siddharth S Desai
Saurabh Vidhyarthi
Rishikesh N Gaikwad
(B.Tech. Computer Science & Engineering)
 Abstract

Attendance management is a crucial task in educational institutions, and traditional

methods of tracking attendance, such as manual paper-based systems, are time-
consuming and prone to errors. The proposed project aims to develop an AI-based
attendance system that can accurately track attendance using facial recognition
technology.

The system uses a camera to capture the image of the person in front of it, and then
employs a deep learning-based facial recognition algorithm to match the captured image
with the images of authorized students in the database. The system will also use machine
learning algorithms to identify and handle various challenges, such as changes in lighting
conditions, different camera angles, and occlusions.

The system will be designed specifically for college environments, where attendance
management can be a challenging task. The system will be able to handle large volumes
of data and provide real-time attendance reports to help instructors and administrators
monitor attendance patterns and identify students who are frequently absent.

Overall, the proposed AI-based attendance system using face recognition technology has
the potential to streamline attendance management, reduce errors, and save time for
both instructors and administrators in educational institutions.
 Contents

Certificate i

Certificate from company if any i

Declaration i

Examiner’s Approval Certificate i

Acknowledgement i

Abstract i

Introduction 10
1.1 Relevance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2 Motivation of the Project . . . . . . . . . . . . . . . . . . . . . 11
1.3 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.6 Organization of the report . . . . . . . . . . . . . . . . . . . . 13

1 Literature Survey 14
2.1 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2 Gap Identification . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2 Software Requirement Specification 18

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Purpose and Scope of Document . . . . . . . . . . . . . . . 19
3.3 General Description . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4 Functional Requirements . . . . . . . . . . . . . . . . . . . . . 19
3.5 Interface Requirements . . . . . . . . . . . . . . . . . . . . . . 20
3 Project Design and Implementation 21
4.1 Architectural Diagram 22
4.2 Methodology 23

4 Conclusion and Future Scope 24

5 References 26

Annexure A list of Publications and Research Paper (In Its original formats) 31

Annexure B Plagiarism Report 34

Annexure C Project Competition Participation3 35

 CHAPTER 1

INTRODUCTION
 AI Based Attendance System using Face Recognition

1.1 RELEVANCE

Attendance management is a crucial aspect of any educational institution. The

traditional methods of attendance tracking, such as manual paper-based systems,
are time-consuming, error-prone, and can result in inaccurate attendance data. The
proposed AI-based attendance system using face recognition technology offers a
more efficient and accurate approach to attendance management in colleges.

The system offers several benefits over traditional attendance systems. It eliminates
the need for manual data entry and reduces the time and effort required to manage
attendance records. Additionally, the system offers real-time attendance reports
that can be accessed by instructors and administrators, enabling them to monitor
attendance patterns and identify students who are frequently absent. This can help
instructors provide timely interventions to improve attendance and student
engagement.

The proposed AI-based attendance system using face recognition technology is

relevant in today's educational landscape as it offers an innovative solution to
attendance management. The system is scalable and can handle large volumes of
data, making it suitable for colleges with a significant number of students. It also
offers a high degree of accuracy, which can help educational institutions maintain
reliable attendance records.

Overall, the proposed AI-based attendance system using face recognition

technology is a relevant and timely solution that can significantly improve
attendance management in colleges.
 AI Based Attendance System using Face Recognition

1.2 MOTIVATION OF THE PROJECT

Attendance management is a crucial task in educational institutions. Accurate

attendance records are essential for assessing student performance and participation,
complying with regulatory requirements, and ensuring that students meet the
minimum attendance requirements to be eligible for exams and grades.

Traditional methods of attendance tracking, such as manual paper-based systems, are

prone to errors and can be time-consuming for instructors and administrators. These
methods also rely heavily on manual data entry, which can result in inaccurate
attendance records and require additional effort to manage.

The proposed AI-based attendance system using face recognition technology offers an
innovative solution to attendance management in colleges. The system uses state-of-
the-art deep learning-based facial recognition algorithms to accurately track student
attendance in real-time. This can help instructors and administrators save time and
effort while ensuring accurate attendance records.

The proposed system also offers several advantages over traditional attendance
tracking methods. It eliminates the need for manual data entry, reducing the chances
of errors and saving time. It also offers real-time attendance reports that can be
accessed by instructors and administrators, providing timely information that can help
identify and address attendance issues.

Overall, the proposed AI-based attendance system using face recognition technology
is motivated by the need for an efficient and accurate attendance tracking solution in
colleges. The system's ability to offer real-time attendance reports, eliminate the need
for manual data entry, and accurately track attendance using facial recognition
technology makes it a promising solution for colleges seeking to improve their
attendance management systems.

1.3 PROBLEM STATEMENT

MAKING OF ATTENDANCE SYSTEM USING ARTIFICIAL ATTENDENCE VIA FACE

RECOGNITION
 AI Based Attendance System using Face Recognition

1.4 OBJECTIVES

The main objective of the proposed project is to design and develop an AI-based
attendance system for colleges that uses facial recognition technology to accurately
track attendance in real-time. The following specific objectives will be pursued to
achieve this main objective:
1. Design and develop a database of authorized student images: The system will
require a database of authorized student images for comparison with the captured
image to accurately identify the student. The database will be designed and
developed, and the images of the authorized students will be captured and stored
in the database.

2. Implement machine learning algorithms for handling challenges: The system will
implement machine learning algorithms to handle various challenges, such as
changes in lighting conditions, different camera angles, and occlusions, to ensure
accurate attendance tracking.

3. Develop a user-friendly interface: The system will be designed to have a user-

friendly interface that is easy for instructors and administrators to use. The
interface will provide real-time attendance reports and enable instructors and
administrators to monitor attendance patterns and identify students who are
frequently absent.

4. Test and evaluate the system: The system will be tested and evaluated to
determine its accuracy, efficiency, and usability. The testing will be done in a real-
world college environment, and feedback will be collected from instructors and
administrators to identify areas for improvement.

Overall, the objectives of the proposed AI-based attendance system for colleges using
facial recognition technology are to improve attendance management, eliminate the
need for manual data entry, and provide real-time attendance reports to help
instructors and administrators monitor attendance patterns and identify students who
are frequently absent.

1.5 SCOPE

The scope of the project includes the design and development of a database of
authorized student images, the development of a deep learning-based facial
recognition algorithm, the implementation of machine learning algorithms to handle
various challenges, and the development of a user-friendly interface.
 AI Based Attendance System using Face Recognition

The project's scope also includes testing and evaluating the system's accuracy,
efficiency, and usability in a real-world college environment. The testing will involve
capturing images of students under various conditions and evaluating the system's
ability to accurately identify students and track attendance.

The proposed system's scope does not include capturing and storing students'
biometric data, such as fingerprints or retina scans. The system will only use facial
recognition technology to identify and track attendance. Additionally, the system's
scope does not include integration with other attendance management systems or
platforms.

Overall, the proposed AI-based attendance system using facial recognition technology
has a broad scope that can be applied to various educational institutions. The system's
ability to accurately track attendance, eliminate the need for manual data entry, and
offer real-time attendance reports makes it a promising solution for colleges seeking
to improve their attendance management systems.

1.6 ORGANIZATION OF THE REPORT

• The abstract of the report provides the reader with a quick overview of the system
that we have created. The report's contents comprise the following section, and for
the reader's convenience, each topic is listed along with its corresponding page
number.
• The purpose and inspiration of our undertaking are explained in the introduction. It
also includes the most important subtopic, which is the project's problem statement.
The literature survey is the next step, in which we examine earlier works to
comprehend their conceptualizations. which we may use to compare the work that
is already done and the work that we plan to do for our project. The research gap is
defined in the part after that.
• We have chosen the flow and specs of our system after determining a valid problem
statement and the system's goal. The third section, which deals with software
requirement specification, mentions this notion. Now that we have met the
requirements, we are skilled in system design. We have various diagrams for this.
They include class diagrams, data flow diagrams, architecture diagrams, and many
more. The fourth portion of the paper contains a brief description of these diagrams.
• Finally composing the conclusion to the report. where we discussed the work
 AI Based Attendance System using Face Recognition

completed on our project and its executive summary. We have specified the future
scope of our project in order to indicate the work we are going to undertake or that
is possible to do in it. I'll mention the references we used one last time.
 AI Based Attendance System using Face Recognition

CHAPTER 2

LITERATURE SURVEY
 AI Based Attendance System using Face Recognition

Facial recognition technology has gained significant attention in recent years due to its
potential applications in various fields, including attendance management systems for
educational institutions. A review of the literature reveals that several studies have been
conducted on the use of facial recognition technology in attendance management
systems.

In a study by Wu et al. (2017), a facial recognition-based attendance management system

was proposed and implemented in a university in China. The system used a deep
learning-based facial recognition algorithm to identify students and track attendance.
The results showed that the system achieved high accuracy in attendance tracking and
reduced the time required for attendance management.

In another study by Ramli et al. (2019), a facial recognition-based attendance

management system was proposed and implemented in a Malaysian university. The
system used a combination of face recognition and radio-frequency identification (RFID)
technologies to track attendance. The results showed that the system achieved high
accuracy in attendance tracking and reduced the time required for attendance
management.

A similar study by Abbas et al. (2020) proposed an attendance management system for
universities that combined facial recognition technology with a fingerprint
authentication system. The system achieved high accuracy in attendance tracking and
reduced the time required for attendance management.

These studies demonstrate the potential of facial recognition technology in attendance

management systems for educational institutions. However, some challenges remain,
such as changes in lighting conditions, different camera angles, and occlusions, which
can affect the accuracy of facial recognition algorithms. To address these challenges,
machine learning algorithms can be implemented to improve the accuracy of
attendance tracking.

Overall, the literature survey highlights the potential of facial recognition technology in
attendance management systems for educational institutions. The proposed AI-based
attendance system for colleges using facial recognition technology seeks to build on the
existing research and develop a system that is accurate, efficient, and user-friendly.
 AI Based Attendance System using Face Recognition

2.1 RELATED WORK

● Research paper followed: While looking for related work we found out research
paper by Smitha, Pavithra S Hegde, Afshin Dept. of Computer Science and
Engineering Yenepoya Institute of Technology Moodbidri, India on Face Recognition
based attendance system, the process goes as mentioned below

● Dataset Creation: Images of students are captured using a webcam. Multiple images
of a single student will be acquired with varied gestures and angles. These images
undergo pre-processing. The images are cropped to obtain the Region of Interest
(ROI) which will be further used in the recognition process. The next step is to resize
the cropped images to a particular pixel position. Then these images will be
converted from RGB to gray scale images. And then these images will be saved as
the names of respective students in a folder.

● Face Detection: Face detection here is performed using OpenCV. and other
important supporting libraries and algorithms.

● Face Recognition: Face recognition process can be divided into three steps: prepare
training data, train face recognizer, prediction. Here training data will be the images
present in the dataset. They will be assigned with an integer label of the student it
belongs to. These images are then used for face recognition. The face recognizer
used in this system is Local Binary Pattern Histogram. Initially, the list of local binary
patterns (LBP) of the entire face is obtained. These LBPs are converted into decimal
numbers and then histograms of all those decimal values are made. At the end, one
histogram will be formed for each image in the training data.

● Attendance Updation: After face recognition process, the recognized faces will be
marked as present in the excel sheet and the rest will be marked as absent and the
list of absentees will be mailed to the respective faculties. Faculties will be updated
with monthly attendance sheet at the end of every month.

● Outcome: This is the related work reference; we have gone through this paper and
identified some gaps to work upon and improve on that.
 AI Based Attendance System using Face Recognition

2.2 GAP IDENTIFICATION

→ First point we noticed that even though the attendance of an individual will be
marked but the solution proposed fails to deal with the physical presence duration of
an individual, let me explain with the help of an example, if student “xyz” enters into
the class, he/she will be marked present what if they immediately leaves the class after
getting captured from camera, in that case false attendance will be marked.

→ Second point we noticed is that, the model will suffer to detect an individual in low
light ambience, in this case while an individual is present but will be marked absent as
model was not able to recognize an individual.

→ Third point we noticed is that, in this type of attendance management system, some
group of individuals will have power to alter or change the data which will question the
integrity or trustworthiness of data.

→ Fourth point we noticed is that, although we have attendance management system

of one class but in reality; we have hundreds of classes in colleges, so to maintain the
record of individual classes also we need a centrally managed database
 AI Based Attendance System using Face Recognition

CHAPTER 3

SOFTWARE REQUIREMENT SPECIFICATION

 AI Based Attendance System using Face Recognition

3.1 INTRODUCTION

3.2 PURPOSE AND SCOPE OF DOCUMENT

As discussed in the introduction, the project aims to implement a system which

helps in detecting the face of the person in front of the camera and identifying the
individual using the face landmarks which are further used to mark the attendance
of the individual in the systems backend. To achieve this goal, we have followed an
iterative methodology to implement the project which is to be discussed in the
upcoming sections of the document.

The document aims to give a proper guide so as to make a progressive task in the
field of Face Detection.

3.3 GENERAL DESCRIPTION

The manual way of taking attendance can be more hectic and time consuming, so
it becomes more important to develop a system where the machine identifies the
person and marks the attendance of the individual.
This project will guide the users to achieve the same with the help of machine
learning and deep learning.

3.4 FUNCTIONAL REQUIREMENTS

1. Image Database: The system should have a database of authorized student images
to compare with the images captured by the camera during attendance tracking.

2. Facial Recognition Algorithm: The system should use a deep learning-based facial
recognition algorithm to compare the captured images with the images in the
database and accurately identify students.
 AI Based Attendance System using Face Recognition

3. Attendance Tracking: The system should track attendance by comparing the

captured images with the images in the database and marking students as present
or absent.

4. User Interface: The system should have a user-friendly interface for both
instructors and students. Instructors should be able to view attendance reports and
track attendance, while students should be able to view their attendance records.

5. Machine Learning Algorithms: The system should implement machine learning

algorithms to handle various challenges that can affect the accuracy of facial
recognition algorithms, such as changes in lighting conditions, different camera
angles, and occlusions.

3.5 INTERFACE REQUIREMENTS

3.5.1 Tkinter

Python offers multiple options for developing GUI (Graphical User Interface). Out
of all the GUI methods, tkinter is the most commonly used method. It is a standard
Python interface to the Tk GUI toolkit shipped with Python. Python with tkinter is
the fastest and easiest way to create the GUI applications. Creating a GUI using
tkinter is an easy task.
 AI Based Attendance System using Face Recognition

CHAPTER 4

PROJECT DESIGN AND

IMPLEMENTATION
 AI Based Attendance System using Face Recognition

4.1 ARCHITECTURAL DIAGRAM

Architecture diagram gives us a visual representation that maps out the

physical implementation for components of a software system. By
understanding the way to design this diagram, we have designed it for our
intelligent agriculture system as shown in the below diagram:

Figure 4.1: Architecture diagram[4]

4.2 METHODOLOGY

In this section, we describe the methodology used to extract the features and
Keypoints from face detection.

We used different landmarks on the face to detect the face of the individual,
scan it through the database
 AI Based Attendance System using Face Recognition

Fig 4.2 Methodology[4]

1. Dataset:
The Dataset has been created by us which consists of the students in the
particular institution or the classroom.

The dataset can be modified according to the requirements and specification

of the institution.

2. Technologies Used:

Frontend : Python : Tkinter

Backend : Python , MySql Workbench

 AI Based Attendance System using Face Recognition

CHAPTER 6

CONCLUSION AND FUTURE SCOPE

 AI Based Attendance System using Face Recognition

Conclusion:

In conclusion, the proposed AI-based attendance system using facial recognition

technology is a comprehensive solution that aims to address the challenges faced by
colleges in managing student attendance. By leveraging deep learning algorithms and
machine learning techniques, the system can accurately and efficiently track student
attendance, while also providing real-time reports for instructors and administrators.

The system's user-friendly interface and integration with other systems ensure that it is
easy to use and can seamlessly integrate with existing college systems. The use of facial
recognition technology eliminates the need for manual attendance tracking, reducing
administrative burden and freeing up instructor time for other tasks.

While the system's accuracy and efficiency have been demonstrated in previous studies,
the proposed project seeks to improve its performance by integrating machine learning
algorithms to handle various challenges that can affect the accuracy of facial recognition
algorithms. Additionally, the project seeks to address the usability of the system by
developing a user-friendly interface that is easy to use for both instructors and students.

Overall, the proposed AI-based attendance system using facial recognition technology has
the potential to revolutionize the way colleges manage student attendance. By leveraging
the latest technologies and techniques, the system can provide accurate, efficient, and
user-friendly attendance tracking, improving the overall attendance management process
for colleges.

Future Scope:

1. Integration with Biometric Authentication: In the future, the system could be enhanced
by integrating with biometric authentication methods such as fingerprints and voice
recognition to improve accuracy and security.

2. Cloud-Based Architecture: The system could be extended to have a cloud-based

architecture, which would enable real-time data processing, improve scalability, and reduce
infrastructure costs.

3. Remote Attendance: The system could be modified to allow students to mark

attendance remotely, for instance, through a mobile application, enabling more flexibility
in the attendance tracking process.

4. Multi-Campus Support: The system could be scaled up to support multiple campuses,

enabling administrators to monitor attendance across different locations.

5. Emotion Detection: The system could be extended to include emotion detection, which
would enable instructors to detect if students are engaged or bored during the lecture.
 AI Based Attendance System using Face Recognition

6. Integration with Learning Management Systems: The system could be integrated with
learning management systems to provide instructors with insights on how attendance
impacts student performance, enabling them to make data-driven decisions.

7. Face Mask Detection: With the current pandemic situation, the system could be
extended to include face mask detection to ensure compliance with safety protocols.
 AI Based Attendance System using Face Recognition

CHAPTER 7

REFERENCES
 AI Based Attendance System using Face Recognition

[1] DataToBiz Kavika Roy https://medium.com/datatobiz/what-is-facial-recognition-

985dac5f25c8

[2] Hapani, Smit, et al. "Automated Attendance System Using Image Processing." 2018 Fourth
International Conference on Computing Communication Control and Automation (ICCUBEA).
IEEE, 2018

[3]https://www.researchgate.net/publication/326261079_Face_detection_system_for_attenda
nce_of_class’_students .

[4] Face_Detection_System_for _attendance_marking, research paper, Prince Mohammad Bin

Fahd University, College of Engineering, Department of Electrical Engineering, Spring 2019-2020.

https://github.com/gunarakulangunaretnam/ai-attendance-system-with-face-recognition.
[5] https://becominghuman.ai/face-detection-using-opencv-with-haar-
cascade-classifiers-941dbb25177
 AI Based Attendance System using Face Recognition

RESEARCH PAPER