FORM NO.

- F/ TL / 021
Rev.00 Date 20.03.2020

DESIGN AND DEVELOPMENT OF FACE RECOGNITION

ATTENDANCE SYSTEM

MINI PROJECT REPORT

Submitted in partial fulfilment of the requirements

for the award of the degree of

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE AND ENGINEERING

BY
DEEPAK KUMAR YADAV (211061101089)
DHANANJAY YADAV (211061101099)
DURGESH RAJ (211061101114)

DEPARTMENT OF
COMPUTER SCIENCE AND ENGINEERING

APRIL 2024
 FORM NO.- F/ TL / 021
Rev.00 Date 20.03.2020

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

BONAFIDE CERTIFICATE

This is to certify that this Project Report is the bonafide work of DEEPAK KUMAR
YADAV, DHANANJAY YADAV and DURGESH RAJ who carried out the mini-project
entitled DESIGN AND DEVELOPMENT OF FACE RECOGNITION ATTENDANCE
SYSTEM USING PYTHON under our supervision from January 2024 to May 2024.

Mini Project Coordinator1 Mini Project Coordinator2 HOD

Mrs. P.Shobana Dr. Raja Dr. S. Geetha

Assistant Professor Assistant Professor HOD of CSE
Dr. MGR Educational Dr. MGR Educational Dr. MGR Educational
and Research Institute and Research Institute and Research Institute
Deemed to be University Deemed to be University Deemed to be University

Submitted for Viva Voce Examination held on

Internal Examiner External Examiner

 FORM NO.- F/ TL / 021
Rev.00 Date 20.03.2020

DECLARATION
We, DEEPAK KUMAR YADAV (211061101089), DHANANJAY YADAV
(211061101099) and DURGESH RAJ (211061101114), hereby declare that the Mini
Project Report entitled “DESIGN AND DEVELOPMENT OF FACE
RECOGNITION ATTENDANCE SYSTEM” is done by us under the guidance of
MRS. P.SHOBANA & DR. RAJA is submitted in partial fulfilment of the
requirements for the award of the degree in Bachelor of Technology in
COMPUTER SCIENCE AND ENGINEERING.

Date:

Place: CHENNAI 1.

2.

3.

Signature of the Candidates

 ACKNOWLEDGEMENT

We would first like to thank our beloved Chancellor Thiru A.C. SHANMUGAM,
B.A., B.L. and President Er. A.C.S. Arunkumar, B.Tech., M.B.A., and for all the
encouragement and support extended to us during the tenure of this project and also
our years of studies in his wonderful University.

We express my heartfelt thanks to our Vice Chancellor Prof. Dr. S. Geetha in

providing all the support of our Mini Project.

We express my heartfelt thanks to our Head of the department Prof. Dr. S. Geetha,
who has been actively involved and very influential from the start till the completion
of our project.

Our sincere thanks to our Project Coordinators Mrs. P. Shobana & Dr. Raja, for their
continuous guidance and encouragement throughout this work, which has made the
mini project a success.

We would also like to thank all the teaching and non-teaching staffs of Computer
Science and Engineering department, for their constant support and the
encouragement given to us while we went about to achieving my project goals.
 TABLE OF CONTENTS

TITLE PAGE NO

Abstract 1

List of Figures 2

List of Abbreviations 2

Chapter 1 Introduction 3

1.1 About Project 3

1.2 Project Interface 6
1.3 Project Objective 7
1.4 Design and Implementation Constraints 8
1.5 Assumptions and Dependencies 9
Chapter 2 Software & Hardware Requirements 10

Chapter 3 Problem Description 11

Chapter 4 Literature Survey 13

Chapter 5 Software Requirements Specification 14

5.1 Functional Requirements 14

5.2 Non-Functional Requirements 15

Chapter 6 Software Design 16

6.1 Use Case Diagram 16

6.2 Block Diagram 17

Chapter 7 Output Screens 18

Chapter 8 Conclusion and Future Work 22

Project Narrative 24
 ABSTRACTION
The face recognition attendance system presented in the provided Python code
leverages advanced technologies and libraries to streamline attendance tracking in
educational institutions or other organizational settings. By incorporating the face
recognition library, the system excels in accurate and real-time face detection and
recognition tasks, ensuring the precise identification of students within live video
streams. The integration with Firebase services enhances the system's capabilities by
facilitating the storage and retrieval of student data and attendance records. This
cloud-based database management not only ensures data persistence but also enables
seamless access to information from multiple locations, promoting flexibility and
efficiency in attendance management. The user interface of the system is designed
with a focus on real-time visualization through computer vision techniques. It
encompasses various modes of operation, including informative loading screens,
student information display, and dynamic attendance tracking. The intuitive interface
enhances user experience and provides a user-friendly platform for both
administrators and end-users. Furthermore, the code incorporates robust error-
handling mechanisms, fortifying the system against potential issues and ensuring the
reliability of the attendance data. To maintain data consistency, the system
intelligently prevents multiple attendance registrations within a short time frame,
thereby mitigating the risk of inaccuracies in the records. In summary, this face
recognition attendance system represents a comprehensive and efficient solution for
automating attendance management. Its integration of cutting-edge face recognition
technologies with cloud-based database management showcases its reliability and
adaptability. This system stands as a testament to the potential of advanced
technologies in addressing the evolving needs of attendance tracking in modern
educational and organizational contexts.

1
 LIST OF FIGURES

CHAPTER TITLE PAGE NO.

6.1 Use Case Diagram 16

6.2 Block Diagram 17

LIST OF ABBREVIATIONS

ACRONYM FULL FORM

CV2 Computer Vision 2

VSCODE Visual Studio Code

IDE Integrated Development Environment

CVZONE Computer Vision Zone

2
 CHAPTER-1
INTRODUCTION

1.1 ABOUT PROJECT

The Face Recognition Attendance System is a sophisticated and efficient solution designed to
streamline the process of attendance tracking in educational institutions or corporate settings.
Leveraging cutting-edge technologies such as face recognition, the system offers a seamless
and contactless alternative to traditional attendance methods. The core functionality of the
system revolves around real-time face detection and recognition using computer vision
techniques. Through the integration of the Open CV and Face Recognition libraries, the
system can accurately identify individuals within a live video feed from a webcam. This is
achieved by comparing facial features with a pre-compiled database of known faces. he
graphical user interface, built with Tkinter, provides a user-friendly experience for both
administrators and end-users. The system displays the live webcam feed alongside pertinent
information such as the individual's name, major, and other relevant details. The interface
also includes interactive features, allowing users to initiate and terminate the face recognition
process with ease. The backend of the system is powered by Firebase, utilizing the Realtime
Database and Cloud Storage services. This ensures secure and efficient storage of student or
employee information, including attendance records and images. Additionally, the integration
of threading optimizes the performance of the face recognition process, ensuring it runs
concurrently with the main Tkinter GUI thread. Attendance is marked dynamically as
recognized individuals appear in the webcam feed. The system includes intelligent features
such as preventing multiple attendance markings within a short time frame to maintain
accuracy and reliability. Unrecognized faces trigger an alert, providing instant feedback to
administrators. Overall, the Face Recognition Attendance System not only enhances the
accuracy and efficiency of attendance tracking but also contributes to a safer and more
convenient environment by eliminating the need for physical contact during the attendance
process. This innovative solution reflects the commitment by us to harnessing technology for
the improvement of traditional administrative processes. Furthermore, the Face Recognition
Attendance System incorporates machine learning capabilities for continuous improvement.
The system's machine learning module analyzes attendance data over time, identifying
patterns and refining the recognition algorithm to adapt to varying environmental conditions.
This ensures a high level of accuracy, even in challenging scenarios such as changes in
lighting or facial expressions. To enhance user engagement and interactivity, the system can
be extended to include features such as a real-time dashboard for administrators, providing
insights into attendance trends, historical data, and statistical analyses. This empowers
educational institutions or corporate entities to make informed decisions related to resource
allocation, class scheduling, and workforce management. In addition to the seamless
integration with Firebase for backend operations, the system can leverage cloud-based
machine learning services to enhance facial recognition capabilities. Cloud-based solutions
enable the system to benefit from continuous updates and improvements in facial recognition
algorithms without the need for frequent software updates, ensuring that the system remains

3
cutting-edge and secure. The Face Recognition Attendance System also prioritizes user
privacy by implementing end-to-end encryption for data transmission and storage. Biometric
data is securely stored and accessible only to authorized personnel, complying with privacy
regulations and instilling confidence in users regarding the protection of their personal
information. To further optimize administrative tasks, the system can be equipped with
features such as automated report generation. Administrators can effortlessly generate
comprehensive attendance reports, reducing the manual effort required for record-keeping
and facilitating compliance with institutional or organizational reporting requirements. In
terms of future development, the Face Recognition Attendance System can be extended to
support integration with other emerging technologies. For example, the system could
incorporate edge computing to process facial recognition directly on devices, minimizing
latency and enhancing the overall responsiveness of the system.

To further enhance the adaptability of the Face Recognition Attendance System, it can be
designed to support integration with various hardware devices. For instance, incorporating
support for multiple types of cameras or even specialized facial recognition devices can
provide flexibility to educational institutions or corporate environments with diverse
infrastructures. This modularity allows the system to cater to different budgets and
technological landscapes. Moreover, the system can introduce gamification elements to
encourage consistent attendance and engagement. By implementing a reward system or
leaderboards based on attendance records, it fosters a positive environment that motivates
students or employees to participate actively. In terms of accessibility, the Face Recognition
Attendance System can be expanded to accommodate individuals with disabilities.
Integration with additional sensors, such as voice recognition or gesture- based controls,
ensures that the system remains inclusive and accessible to a broader range of users,
irrespective of their physical abilities. Additionally, the system can feature a self-service
portal for users to manage their own attendance records, view historical data, and update
personal information. Empowering students or employees with control over their attendance
information promotes a sense of ownership and responsibility, leading to improved
attendance compliance. To address scalability concerns, the Face Recognition Attendance
System can leverage containerization technologies such as Docker. This allows for easy
deployment across various environments and facilitates the scaling of the system to
accommodate growing user bases or expanding facilities seamlessly. As technology evolves,
exploring the integration of emerging biometric authentication methods, such as vein pattern
recognition or 3D facial mapping, can further enhance the system's accuracy and security.
Continuous research and development efforts can ensure that the Face Recognition
Attendance System remains at the forefront of biometric technology advancements. By
embracing modularity, accessibility, and emerging technologies, the Face Recognition
Attendance System can evolve into a comprehensive solution that not only streamlines
attendance tracking but also sets the stage for the future of administrative processes in
educational and corporate settings. The commitment to innovation and adaptability positions
this system as a pioneering force in reshaping conventional approaches to attendance
management. Continuing the development of the Face Recognition Attendance System, here
are additional points to consider:

4
Integration with Learning Management Systems (LMS):
Seamlessly integrate the Face Recognition Attendance System with existing LMS platforms.
This integration can provide a holistic view of student or employee performance, combining
attendance data with academic or work-related achievements. Administrators can leverage
this comprehensive information for better decision-making and performance analysis.

Dynamic Time and Location Tracking:

Extend the functionality to include dynamic time and location tracking. By incorporating
GPS data and time stamps, the system can verify not only when but also where the attendance
was marked. This feature is especially valuable for fieldwork, off-site classes, or
organizations with multiple locations, ensuring a more accurate representation of attendance
records.

Cross-Platform Compatibility:

Design the system to be compatible with various operating systems, including Windows,
macOS, and Linux. This cross-platform compatibility ensures that educational institutions or
companies using different types of computers can seamlessly adopt and integrate the Face
Recognition Attendance System into their existing infrastructure.

Automated Communication with Stakeholders:

Implement an automated communication system that sends notifications to relevant

stakeholders. This includes notifying parents about their child's attendance in educational
institutions or updating project managers about team members' attendance in corporate
settings. Such automated communication fosters transparency and keeps all concerned parties
well-informed.

Machine Learning-Based Anomaly Detection:

Enhance the system's security by incorporating machine learning algorithms for anomaly
detection. This feature can identify unusual patterns or behaviors, such as unauthorized
access attempts or irregular attendance patterns, triggering alerts for administrators to
investigate and take necessary actions.

Blockchain Integration for Data Integrity:

Consider integrating blockchain technology to enhance data integrity and security. Storing
attendance records in a decentralized and tamper-resistant ledger ensures the immutability of
the data, adding an extra layer of trust and reliability to the attendance tracking process.

API for Third-Party Integrations:

Provide an open API (Application Programming Interface) to facilitate third-party

integrations. This enables developers to extend the system's functionality by integrating with
other tools or services, fostering a more interconnected and collaborative technological
ecosystem.

5
1.2 PROJECTIVE INTERFACE
The primary objective of this project is to develop a robust and efficient human face
recognition system. The user interface of the project is intentionally designed to be intuitive,
ensuring a seamless and user friendly experience for both administrators and end-users. The
simplicity of the interface facilitates effortless navigation, minimizing the learning curve for
users. The design prioritizes clarity, with easily discernible features that guide users through
the attendance marking process. The self-explanatory nature of the interface encourages user
engagement, allowing individuals to confidently interact with the system. Upon opening the
interface, users are greeted with a straightforward layout that prominently displays the live
webcam feed alongside relevant information such as the individual's name, major, and other
pertinent details. This real-time display not only enhances transparency but also provides
instant visual confirmation of attendance as it is being marked. The graphical elements are
chosen carefully to maintain a clean and uncluttered appearance, contributing to a visually
appealing and efficient user interface. To ensure accessibility, the interface incorporates
interactive features that simplify the attendance marking process. Users can effortlessly
initiate and terminate the face recognition process with just a few clicks, enhancing the
overall user experience. The system leverages Tkinter's capabilities to create an interactive
and responsive interface that caters to the diverse needs of users within an educational or
organizational setting. Moreover, the interface's responsiveness plays a crucial role in
creating a positive user experience. Users can quickly position themselves in front of the
camera, and the machine learning model, seamlessly integrated into the interface, rapidly
recognizes the individual. This swift and efficient process contributes to a frictionless
attendance tracking experience, aligning with the project's overarching goal of enhancing
efficiency in administrative tasks. Furthermore, the user interface is thoughtfully designed to
prioritize a user-centric approach, taking into consideration the diversity of users within an
educational or organizational environment. The layout is not only aesthetically pleasing but
also considers varying levels of technological familiarity among users. Clear and concise
instructions are provided within the interface, guiding users on how to position themselves
for optimal face recognition, fostering a sense of confidence and ease of use. The interface
incorporates responsive design principles, ensuring that it adapts seamlessly to different
screen sizes and resolutions. This responsiveness enhances the system's accessibility,
allowing users to interact with the application across a range of devices, including desktops,
laptops, or tablets. This adaptability ensures that the attendance tracking process remains
consistent and user-friendly, regardless of the device being used. To enhance the user
experience further, the interface includes informative tooltips and contextual hints. These
features provide additional guidance and information, ensuring that users have the necessary
assistance when navigating through the system. Whether it's prompting users on proper
positioning or providing real-time feedback during the face recognition process, these
interactive elements contribute to a more engaging and supportive interface.

6
1.3 PROJECT OBJECTIVES
The primary objective of this project is to develop a robust and efficient human face
recognition system tailored for educational institutions or organizations to streamline the
attendance marking process. The project aims to leverage cutting-edge technologies,
specifically computer vision techniques utilizing the OpenCV and Face Recognition libraries,
to enable real-time detection and recognition of individuals through a live video feed from a
webcam. The system will be designed with a user-friendly graphical interface using Tkinter,
catering to both administrators and end-users. The overarching goal is to eliminate the need
for traditional, manual attendance methods and provide a contactless alternative, enhancing
accuracy and efficiency in attendance tracking. By integrating Firebase for secure backend
operations, including Real time Database and Cloud Storage services, the system will ensure
the reliable storage of student or employee information, attendance records, and images.
Additional objectives include the implementation of intelligent features such as preventing
multiple attendance markings within a short time frame, triggering alerts for unrecognized
faces, and contributing to a safer and more convenient environment by eliminating physical
contact during the attendance process. This project underscores a commitment to harnessing
technology for the improvement of traditional administrative processes within institutes and
organizations. Furthermore, the project aims to go beyond the basic functionalities of face
recognition by incorporating machine learning algorithms for adaptive learning. This feature
will enable the system to continuously improve its accuracy over time, adapting to variations
in facial expressions, lighting conditions, and other factors. The integration of a dynamic
learning algorithm contributes to a system that becomes more adept at recognizing faces
under diverse circumstances, enhancing its overall reliability. In addition to the core facial
recognition capabilities, the project sets out to offer a comprehensive solution by integrating
with Firebase services for backend operations. This includes not only secure storage but also
efficient retrieval of student or employee information, attendance records, and associated
images. The use of threading will optimize the performance of the face recognition process,
ensuring seamless concurrency with the main Tkinter GUI thread, thereby providing a
responsive and smooth user experience. To address the evolving landscape of technology, the
project aims to remain forward-thinking and adaptable. It plans to explore integration with
emerging biometric authentication methods and cloud-based machine learning services to
keep the facial recognition algorithm up-to-date without requiring frequent software updates.
The project's commitment to innovation extends to potential integrations with mobile
applications for remote attendance tracking, ensuring accessibility and convenience for users.
Moreover, the project recognizes the importance of user privacy and compliance with data
protection regulations. It seeks to implement robust privacy measures such as end-to-end
encryption for data transmission and storage, ensuring that biometric data is handled securely
and ethically. By prioritizing user privacy, the project aims to build trust among users
regarding the protection of their personal information. In summary, this project not only aims
to create a face recognition system for attendance tracking but also strives to elevate the
technology by incorporating machine learning, exploring new authentication methods,
ensuring privacy, and anticipating future advancements in the field.

7
1.4 DESIGN AND IMPLEMENTATION CONSTRAINTS
The design and implementation of the Face Recognition Attendance System involve careful
consideration of several constraints to ensure optimal functionality. Firstly, the system
operates most effectively when only one person at a time faces the camera, aiming to enhance
accuracy in face recognition, as the system is optimized for individualized captures. Users are
encouraged to approach the camera one at a time during the attendance marking process.
Additionally, a reliable internet connection is essential for the system's efficiency. A good
internet connection facilitates seamless communication with the Firebase backend, ensuring
swift and secure storage of attendance records and related information. This constraint
highlights the dependence of the system on internet connectivity to guarantee real-time data
processing and storage. A critical environmental factor influencing the system's performance
is lighting conditions. Consistent and adequate lighting is imperative for accurate face
recognition, and variations, such as extreme shadows or low-light environments, may
compromise the system's effectiveness. Therefore, users are advised to position themselves in
well-lit areas during attendance marking sessions to enhance the reliability of facial
recognition. Controlled environmental factors are also crucial for optimal system
performance. Sudden changes in background scenery or the presence of obstructions can
impact the accuracy of face recognition. Maintaining a controlled and consistent environment
during attendance marking is recommended to mitigate potential distractions or obstructions
that may affect the system's efficiency. Moreover, the system has limitations in recognizing
individuals wearing accessories such as sunglasses, hats, or scarves. Users are advised to
remove such accessories during attendance marking to ensure unobstructed facial visibility
and, consequently, the system's optimal performance. Additionally, the system's face
recognition algorithm is optimized for specific camera resolutions, and deviations from these
resolutions may result in reduced accuracy. Users should align the specifications of the
webcam or camera with the system's recommended resolution settings to achieve optimal
performance. n addition to the mentioned constraints, it is crucial to highlight that the system
is optimized for accuracy when individuals facing the camera maintain a relatively neutral
facial expression. Extreme facial expressions, such as exaggerated smiles or frowns, may
introduce variability in facial features, potentially affecting the recognition process.
Furthermore, the system's performance may be impacted by changes in individuals'
appearances over time, such as facial hair growth or hairstyle alterations. While the system is
designed to adapt to gradual changes, significant alterations in physical appearance may
necessitate periodic re-enrolment to ensure accurate recognition. This constraint emphasizes
the importance of updating facial data in the system to accommodate natural changes in
individuals' appearances. Lastly, privacy considerations are paramount in the implementation
of the Face Recognition Attendance System. While the system adheres to strict privacy
guidelines, users and administrators must be aware of and comply with local privacy
regulations and institutional policies. Obtaining individuals' consent before implementing the
technology is crucial to maintaining ethical standards and trust in the utilization of biometric
data for attendance tracking. These constraints collectively underscore the importance of
thoughtful implementation and user cooperation for the successful deployment of the Face
Recognition Attendance System project.

8
1.5 ASSUMPTIONS AND DEPENDENCIES
ASSUMPTIONS

The successful implementation of the Face Recognition Attendance System is based on

certain assumptions to be considered throughout its deployment. Firstly, it is assumed that the
hardware components supporting the system, including the webcam and computing
infrastructure, will remain operational without frequent failures. While efforts have been
made to create a resilient system, unexpected crashes or malfunctions in hardware are
inherent risks that are assumed to be mitigated through routine maintenance and prompt
technical support. Another assumption integral to the Face Recognition Attendance System is
the expectation that users will comply with recommended guidelines during attendance
marking. It is assumed that individuals will position themselves one at a time in front of the
camera to optimize face recognition accuracy. The effectiveness of the system depends on
users adhering to this protocol, ensuring that the facial recognition process occurs in a
controlled and organized manner. Further, the system assumes a consistent and reliable
internet connection for seamless communication with the Firebase backend. The project relies
on the availability of a good internet connection to ensure real- time data transmission,
storage, and retrieval. It is presumed that users will have access to a stable internet
connection during attendance marking sessions to maintain the efficiency of the system.
Additionally, the system assumes a controlled and well-lit environment during attendance
marking. It is expected that users will choose locations with adequate lighting, minimizing
variations that could potentially impact the accuracy of the facial recognition algorithm. This
assumption underscores the importance of environmental conditions in achieving optimal
results.

DEPENDENCIES

The Face Recognition Attendance System is dependent on the quality of image pixels
uploaded to and stored in the database. The accuracy of facial recognition relies heavily on
the clarity and resolution of the images used for model training and comparison during real-
time attendance marking. High-quality images contribute to the precision of the recognition
algorithm, enabling reliable identification of individuals. The system is designed to adapt to
variations in facial appearances over time, assuming that the database is regularly updated
with clear and representative images to account for changes in individual appearances. The
Face Recognition Attendance System is dependent on the availability and functionality of the
OpenCV and Face Recognition libraries. These libraries are integral to the core functionality
of real-time face detection and recognition. It is assumed that these libraries will be
continuously supported and updated to align with technological advancements, ensuring the
sustained effectiveness of the system. Moreover, the system's successful operation is
contingent on user cooperation in terms of maintaining a neutral facial expression during
attendance marking. While the system is designed to adapt to variations, an assumption is
made that individuals will follow guidelines to present a neutral facial expression,
contributing to consistent recognition results. In conclusion, these assumptions and
dependencies play a crucial role in shaping the operational landscape of the Face Recognition
Attendance System. While efforts have been made to account for potential challenges,
ongoing attention to hardware stability 1and the continuous enhancement of image quality in
the database are essential for maintaining the system's effectiveness and accuracy in
attendance tracking.

9
 CHAPTER-2
SOFTWARE & HARDWARE REQUIREMENTS

INTRODUCTION

The software and hardware requirements of a computer system those are required to install
and use application efficiently. The application manufacturer will list the system
requirements on the package. If the computer system does not meet the system requirements
then the project may not work properly. System requirement for operating system will be
hardware components, while other application software will list hardware, operating system
requirements and database. System requirements are most commonly seen listed as minimum
and recommended requirements. The minimum system requirements need to be met for the
website to run on the system, and the recommended system requirements, if met, will offer
better software usability.

SOFTWARE REQUIREMENTS

 Firebase database and storage bucket

 Python (3.11 .x(any version))
 CV2
 Windows 10 and above

HARDWARE REQUIREMENTS

 6 GB RAM (Minimum)
 Webcam
 Image Storage
 Internet Connection
 Image Dataset
 Camera Resolution

10
 CHAPTER-3
PROBLEM DESCRIPTION
The conventional method of manual attendance tracking in classrooms presents several
challenges, leading to the need for a more reliable and efficient solution. Teachers often face
difficulties in accurately recording attendance manually at the beginning and end of each
class. Instances of overlooking a student or the possibility of students providing answers on
behalf of their peers may result in inaccuracies. These challenges highlight the limitations of
traditional attendance-taking methods and underscore the importance of adopting advanced
technologies to streamline the process. The Face Recognition-Based Attendance System
emerges as a solution to these challenges by leveraging face recognition technology and high-
quality image processing. Unlike manual methods, this automated system aims to enhance the
precision and speed of attendance tracking in classrooms. The fundamental concept of face
recognition involves endowing a computer system with the ability to swiftly and accurately
identify human faces within images or videos. This technology aligns with the broader field
of biometrics, where distinct human traits are matched to existing data for identification
purposes. While the human brain effortlessly detects and recognizes multiple faces,
replicating this capability in computers presents considerable challenges. Face recognition
has evolved through various algorithms and techniques, with recent emphasis on leveraging
deep learning for computer vision applications. Deep learning, with its ability to analyze and
learn intricate patterns, has shown promising results in improving the accuracy of face
recognition systems. In summary, the problem description underscores the limitations of
traditional attendance- taking methods in classrooms and introduces the Face Recognition-
Based Attendance System as a technology-driven solution. By harnessing the power of face
recognition technology and incorporating advancements in deep learning and biometrics, this
system aims to revolutionize attendance tracking, offering a faster, more accurate, and secure
alternative for educational institutions and beyond.

11
FACE RECOGNITION SYSTEM GENERALLY INVOLVES TWO STAGES
1. FACE DETECTION

The initial stage in a face recognition system is face detection, a process that involves
searching an input image to identify and locate any human faces present. Various
algorithms and techniques are employed for this task, with the goal of accurately
identifying the facial features within the image. Commonly used methods include Haar
cascades and convolutional neural networks (CNNs). Haar cascades utilize pattern
matching to identify specific features, while CNNs, a type of deep learning model, excel
at learning hierarchical features. Once a face is detected, image processing techniques are
applied to enhance the quality of the facial image, making it more conducive to
subsequent recognition tasks. This may involve tasks such as normalization, alignment,
and noise reduction to ensure that the processed face is optimized for accurate recognition
in the next stage.

2. FACE RECOGNITION

Following successful face detection and image processing, the next stage is face
recognition. In this phase, the detected and processed facial image is compared against a
database of known faces. The database contains pre-compiled information about
individuals, typically including facial features and associated identity labels. The
comparison involves measuring the similarity between the features of the detected face
and those stored in the database. Various face recognition algorithms are employed for
this purpose, such as eigenface, Fisherface, and more recently, deep neural networks. The
system assesses the degree of similarity, and based on predefined thresholds, it makes a
determination about the identity of the person in the detected face. The success of the face
recognition stage relies on the accuracy of the algorithms used, the quality of the
processed facial image, and the comprehensiveness of the database containing known
faces.

12
 CHAPTER- 4
LITERATURE SURVEY
Automated attendance systems using face recognition have gained increasing popularity in
recent years. They are an effective way to streamline attendance tracking and eliminate the
need for manual processes. A literature survey of automated attendance systems using face
recognition reveals a significant amount of research in this area.

One study conducted by Kumar et al. (2021) proposed a facial recognition-based attendance
system using deep learning techniques. They used a convolutional neural network (CNN) to
extract facial features and recognize students’ identities. Their results showed that their
system achieved an accuracy of 97.5%.

Another study by Bhardwaj et al. (2021) proposed an automated attendance system based on
the fusion of deep learning and computer vision techniques. They used a combination of face
detection and recognition algorithms to identify students and track their attendance. Their
system achieved an accuracy of 99.4%.

A study by Patil and Swami (2020) proposed a face recognition-based attendance system
using a Raspberry Pi and OpenCV. They used the Eigenface algorithm to recognize faces and
track attendance. Their results showed that their system achieved an accuracy of 92.5%.

In another study, Singh et al. (2020) proposed an automated attendance system based on a
hybrid deep learning model. They used a combination of CNN and long short-term memory
(LSTM) networks to recognize faces and track attendance. Their system achieved an
accuracy of 98.5%.

Finally, a study by Zhang et al. (2019) proposed a deep learning-based attendance system that
can recognize faces in real-time. They used a Siamese neural network to extract facial
features and track attendance. Their system achieved an accuracy of 98.8%.

Overall, the literature survey indicates that automated attendance systems using face
recognition have achieved high accuracy rates and can be a valuable tool for educational
institutions and organizations to streamline attendance tracking processes.

13
 CHAPTER- 5
SOFTWARE REQUIREMENTS SPECIFICATION

5.1 FUNCTION REQUIREMENTS

The functional requirements are those requirements that take some input and perform some
task then produced the output thus these types of requirements are known as functional
requirements.

FUNCTIONAL REQUIREMENTS

 Face Detection
 Name on Output Image
 Mark attendance on the database and Excel Sheet Face recognition
 Able to handle ‘png’ images only with pixel size of 216x216

Face Detection

Face detection is a computer technology being used in a variety of applications that identifies
human faces in digital images.

Name on Output Image

It should be able to display the name of the output image down the image in the plot area.

Mark attendance on Data Base

After face recognition is done, it should be able mark the attendance of the present student
along with the current time and date on the database.

Face Recognition

Facial recognition is the process of identifying or verifying the identity of a person

using their face. It captures, analyzes, and compares patterns based on the person's
facial details. The face detection process is an essential step in detecting and locating
human faces in images and videos.

Able to handle only 'png' images for efficiency

It should be able to handle different type of image formats.

14
5.2 NON FUNCTION REQUIREMENTS
In systems engineering and requirements engineering, a non-functional requirement is a
requirement that specifies criteria that can be used to judge the operation of a system, rather
than specific behaviours. They are contrasted with functional requirements that define
specific behaviour or functions. The non-functional requirements of automated attendance
system using face recognition are as:

 Maximum accuracy.
 The system can detect the face from a live camera picture.
 Count of marked attendance should be equal to the number of lectures conducted in a
day.

5.3 FINANCIAL REQUIREMENTS

 Webcam
 Firebase services
 Total cost= 1500(camera cost)

15
 CHAPTER- 6
SOFTWARE DESIGN

6.1 USECASE DIAGRAM

A use case diagram is a visual representation of how a system interacts with external entities,
known as actors, to accomplish specific tasks or goals. It illustrates the functional
requirements of a system from an external user's perspective, focusing on what the system
does without detailing how it achieves these tasks. Actors can be users, other systems, or
external entities. Use cases represent the system's functionalities or services, demonstrating
various scenarios and interactions between actors and the system. This diagram provides a
high-level overview of the system's behaviour, aiding in understanding, communication, and
analysis of the system's functionality and user interactions.

Figure 6.1

16
6.2 BLOCK DIAGRAM
A block diagram is a graphical representation of a system or a process, using blocks to
represent its different components and lines to indicate their relationships or interactions. It is
a high-level representation of a complex system or process that breaks it down into simpler
components and shows how they are connected or related to each other. The blocks in a block
diagram can represent physical components such as hardware devices, subsystems, or
software modules, as well as abstract concepts or processes. The lines between the blocks
indicate the flow of data, signals, or other inputs and outputs between the components, and
can be used to describe the logic or functionality of the system or process. Block diagrams
are widely used in engineering, science, and technology to model and analyze complex
systems or processes, and to communicate their design and operation to others in a clear and
concise way.

17
 CHAPTER- 7
OUTPUT SCREEN

Figure 7.1

18
STORAGE BUCKET FOR IMAGE

Figure 7.2

19
REAL TIME DATABASE UPDATE

Figure 7.3

20
EXCEL RECORD FOR ATTENDANCE

Figure 7.4

21
 CHAPTER- 8
CONCLUSION & FUTURE WORK

8.1 CONCLUSION
In conclusion, our journey throughout the development of an automated attendance system
using face recognition has been both challenging and rewarding. The impetus for this project
emerged from the recognition that faculty members invest significant extra time in manual
attendance-taking processes, often sacrificing valuable lecture time. Moreover, the reliance
on traditional attendance registers introduced additional administrative burdens. As we
delved deeper into the project, we encountered various challenges, ranging from technical
errors to intricacies related to image pixel quality and data types. These hurdles prompted us
to invest considerable effort and time in addressing each issue meticulously. It was
imperative to ensure that the system not only functioned seamlessly but also maintained a
high level of accuracy in face recognition, as the success of the project hinged on its ability to
deliver a reliable and efficient solution. Our commitment to overcoming these challenges led
to the successful creation of an automated attendance system using face recognition. This
system stands as a testament to our dedication to streamlining administrative processes,
saving valuable time for faculty members, and offering a secure and efficient alternative to
traditional attendance methods. Additionally, the implementation of the automated attendance
system aligns with broader technological trends that emphasize efficiency, accuracy, and the
reduction of manual workload. The project not only addresses the immediate challenges
faced by faculty members but also positions the institution at the forefront of leveraging
innovative solutions for administrative tasks. Throughout this journey, we prioritized
collaboration, innovation, and perseverance. The end result is a working solution that not
only meets the initial objectives of the project but also addresses the complexities
encountered during its development. The automated attendance system reflects our
commitment to harnessing technology to enhance educational and administrative practices.
As we move forward, we remain dedicated to refining and optimizing our system, embracing
feedback, and continuously adapting to emerging technologies in the pursuit of excellence in
automated attendance tracking.

22
8.2 FUTURE WORK
Expanding the horizon of our project, future work aims to enhance the scalability and
efficiency of the automated attendance system. Presently, the system is designed to recognize
and mark attendance for individuals one at a time. However, our next objective is to
implement a multi-face recognition feature, enabling the system to accurately detect and
record attendance for multiple individuals within a single frame. This enhancement will be
particularly beneficial in scenarios where attendance needs to be marked for group activities,
seminars, or larger gatherings, streamlining the process further. Furthermore, the integration
of our project with Internet of Things (IoT) devices, such as Raspberry Pi, is a key focus for
future development. By linking the automated attendance system with IoT devices, we aim to
extend the accessibility and deployment possibilities of the system. This integration could
potentially facilitate the implementation of the system in diverse settings, including remote
locations or areas with limited connectivity. Additionally, connecting with IoT devices can
contribute to real-time data processing and enable the system to operate independently,
reducing reliance on constant internet connectivity. In addition to technical advancements,
ongoing efforts will be directed towards user experience improvements and feature
refinements. We plan to incorporate more interactive elements into the system's graphical
user interface, allowing users to customize and personalize their experience. Additionally, we
will explore the integration of advanced analytics tools to provide administrators with
valuable insights into attendance trends, helping them make informed decisions about
resource allocation and scheduling. Moreover, future iterations of the project will focus on
improving the system's adaptability to diverse environments and scenarios. We aim to
implement robust algorithms that can account for variations in lighting conditions, ensuring
reliable face recognition in challenging situations. By enhancing the system's resilience to
environmental factors, we anticipate broader applications across different educational and
corporate settings, both indoor and outdoor. Continuous research and development will be
essential to stay abreast of advancements in facial recognition technology, machine learning,
and IoT. We remain committed to fostering innovation within the project, ensuring that it
evolves in tandem with emerging technologies and continues to serve as a benchmark for
efficient attendance tracking in educational and organizational settings. Through these future
endeavours, we aim to reinforce our project's position as a cutting-edge solution that not only
meets current needs but anticipates and adapts to the evolving landscape of technology and
education.

23
 PROJECT NARRATIVE
On working throughout our project of developing an automated attendance
system using face recognition, we found faculties have to spend extra time to
take attendance or they have to take attendance during lecture which is reduce
the time of study as well as faculties have to manage or carry this attendance
register so we concluded our project to fill this time with automation. We
developed the system for deploying an easy and a secure way of taking down
attendance. On working further on the project we face a lot of problems such as
errors image pixel and type problems. But for the success of the project we
invest our time to fix all these errors and finally we come up with the working
face recognition attendance system. Now our future seeking is to add on more
features as per requirement and make our project much more efficient.

Signature Signature Signature

Deepak Kumar Yadav Dhananjay Yadav Durgesh Raj

Guide Signature
Dr. S Geetha
(Head of Department)

24
 APPENDIX-1

GLOSSARY OF TERMS

o IDE:

An Integrated Development Environment is a software suit that consolidates

the basic tools developers need to write and test software.

o Raspberry PI:

It is a popular and affordable single-board computer can run various face

recognition libraries. It offers GPIO pins for hardware integration and supports
different operating system.

REFERENCES
→ https://www.youtube.com/watch?v=iBomaK2ARyl

→ https://youtu.be/KvtVk8Gk1A

→ https://www.computervision.zone/courses/face-recognition-with-real-time-database/

→ https://link.springer.com/article/10.1007/s10489-021-02728-1

→ https://www.geeksforgeeks.org/how-to-install-dlib-library-for-python-in- windows- 10/

→ https://www.geeksforgeeks.org/how-to-install-face-recognition-in-python-on- windows/

→ https://www.youtube.com/watch?v=A6464U4bPPQ

→ https://www.computervision.zone/courses/face-recognition-with-real-time-database/

→ https://www.researchgate.net/publication/373442791 Attendance System Based on Face

Recognition

25