ARBAMINCH UNIVERSITY

INSTITUTION OF TECHNOLOGY

FUCALTY OF ELECTRICAL AND COMPUTER

ENGENDERING
Computer engineering stream
Tittle : Face recognition based attendance system
SECTION “A”
Adivsor Name: Iyasu.
 DECLARATION
We declare that this thesis report is original work of ours. We try to develop a system that
recognize the face of person and attendance tracking system. All writings in this document
contain things that are done by our effort and original work carried out by ourselves under
guidance of our advisor. This thesis has not been presented for a degree in any other University
and we confirm that all the source of material used for this thesis have been fully
acknowledged.

Name of the student Signature date

1. Chalew Habte ………… …………
2. Bakana Soressa ………… …………
3. Dengmoses ………… …………
4. Gatcouth Gai ………… …………
5. Kalkidan Iyasu ……….. …………..
Approval
This is to certify that the project titled “loan prediction system for saving and credit association) based
Machine learning language” Carried out by the students listed their name above has been read and
approved for meeting part of the requirements and regulations governing the award of the Bachelor of
Engineering (in Electrical and Computer Engineering under the stream of computer Engineering) degree
at arbaminch University.
Approved by:
Advisor name: Signature date
Mr.Iyasu.D …………… …………
Department head name: signature date
Mesfin .G …………… …………
Examiners names: signature date
…………………… ……………. …………
…………………… ……………. …………
 ACKNOWLEDGEMENT
First of all we would like to thank our God, the almighty for his blessing all the time and
assisting us for doing the thesis project from the beginning to the end. And we would like to
present our gratitude to University of arbaminch which managed and helped us in getting this
degree program. We would like sincerely thank our advisor Mr. Iyasu for his valuable
guidance, consecutive comment, suggestion, and e ment to do this project. We ncourage
would like to express our sincere thanks and all of our department staff members for their
helping hands whenever required. We also extend our gratefulness to our family and the
peoples who are directly or indirectly involved in the successful completion of this thesis
project work. Moreover, we would like to thank Electrical and Computer Engineering
Department instructors of Arba minch University for their kind commitment in the teaching
and learning process throughout the past years. Finally, we would like to thank our Group
members, for our initiation, cooperation and fully participation during this project.
List of Figures
Figure 3.1 : Neural network approach face recognition………………………………………….19

Figure 3.2 : Algorithm of face detection part………………………………………………........20

Figure 3.3 : Image in different white balance …………………………………………………...20

Figure 3.4 : Facial Feature Extractions (Left) and Face Image (Right)……………………......21

Figure 3.5 : Face recognition part………………………………………………………………..22

Figure 3.6 : Block diagram of face recognition attendance system…………………………….24

Figure 3.7 : Biometric face recognition how it work…………………………………………...25

Figure 3.8 : Use case model of face recognition attendance system……………………….…..29

Figure 3.9 : System architecture analysis…………………………………………………..….30

Figure 4.1 : The face tracing attendance system active………………………………………..38

Figure 4.2 : Tracing the face of student………………………………………………………..38

Figure 4.3 : After verifying student face with the data base………………………………......49

Figure 4.4 : The excel file of student attendance list………………………………….………40

Figure 4.4 : Student image in storage…………………………………………………………40

Figure 4.5: The total step of face recognition attendance implementation step in

configuration………………………………………………………………………………………41

Figure 4.6 : Real time database student information………………………………………….42

List of acronym

APIs = Application Programming Interfaces

CCPA = California Consumer Privacy Act

CCTV = Closed-Circuit Television

CNNs = Convolutional Neural Networks

CUDA = Compute Unified Device Architecture

CV2 = OpenCV (Open Source Computer Vision Library)

FFNN = Feedforward Neural Network

FPGAs = Field-Programmable Gate Arrays

GDPR = General Data Protection Regulation

GPUs = Graphics Processing Units

HOG = Histogram of Oriented Gradients

NPUs = Neural Processing Units

SDKs = Software Development Kits

TPUs = Tensor Processing Units

Table of Contents
DECLARATION.........................................................................................................................................ii
Approval....................................................................................................................................................iii
ACKNOWLEDGEMENT...........................................................................................................................iv
List of Figures.........................................................................................................................................v
List of acronym..........................................................................................................................................vi
Abstract...................................................................................................................................................viii
CHAPTER ONE.........................................................................................................................................1
1 Introduction.........................................................................................................................................1
1.1 Background of Study..................................................................................................................1
1.2 Statement of the problem............................................................................................................2
1.3 Objectives...................................................................................................................................3
1.3.1 General objectives...............................................................................................................3
1.3.2 Specific objectives...............................................................................................................3
1.4 Significance of the project...........................................................................................................3
1.5 Scope and Contribution of the Project........................................................................................4
1.6 Limitation of the project...............................................................................................................4
1.7 Thesis Outline.............................................................................................................................5
CHAPTER TWO........................................................................................................................................6
2 Literature Review...............................................................................................................................6
2.1 Literature Survey on the Face Detection and Feature Extraction...............................................6
2.2 Methodology................................................................................................................................6
2.3 Evaluation...................................................................................................................................8
2.3.1 Accuracy..............................................................................................................................8
2.3.2 Precision..............................................................................................................................8
CHAPTER THREE....................................................................................................................................9
3 System Design and Analysis..............................................................................................................9
3.1 Introduction.................................................................................................................................9
3.2 DESIGN OF A FACE RECOGNITION SYSTEM........................................................................9
3.3 Input Part....................................................................................................................................9
3.4 Face Detection Part..................................................................................................................10
3.5 Face Recognition Part..............................................................................................................11
3.6 Working principle......................................................................................................................13
3.7 Simulation design and analysis.................................................................................................14
3.8 System Architecture Design......................................................................................................14
 3.9 Algorithm Selection...................................................................................................................14
3.10 Simulation Setup.......................................................................................................................15
3.10.1 Implementation..................................................................................................................15
3.10.2 Performance Evaluation....................................................................................................15
3.10.3 Validation and Verification.................................................................................................15
3.10.4 Optimization and Iteration..................................................................................................15
3.11 Software and hardware component used.................................................................................15
3.11.1 Hardware component........................................................................................................16
3.11.2 Software requirement........................................................................................................16
3.12 System analysis........................................................................................................................17
3.13 Challenge faced during software and hardware implementation..............................................22
3.14 Measure to overcome the problems.........................................................................................22
3.15 Purpose of the system design...................................................................................................23

Abstract
Face recognition is among the most productive image processing applications and has a pivotal role in
the technical field. Recognition of the human face is an active issue for authentication purposes
specifically in the context of attendance of students. Attendance system using face recognition is a
procedure of recognizing students by using face biostatistics based on the high definition monitoring and
other computer technologies. The development of this system is aimed to accomplish digitization of the
traditional system of taking attendance by calling names and maintaining pen-paper records. Present
strategies for taking attendance are tedious and time-consuming. The primary problem addressed is the
need for a more reliable, efficient, and non-intrusive method to monitor attendance, which can also
prevent fraudulent attendance marking. To solve this problem, the proposed system uses a combination
of CCTV cameras for capturing real-time video feeds and Graphics Processing Units (GPUs) for
accelerated image processing. The result is a highly accurate and efficient attendance tracking system
that can automatically record and store attendance data in real-time. The system demonstrates high
accuracy rates in diverse lighting conditions and environments, significantly reducing the time and effort
required for attendance management. In conclusion, the face recognition-based attendance tracking
system provides a modern solution to traditional attendance recording challenges, offering enhanced
accuracy, efficiency, and security.
 CHAPTER ONE

1 Introduction
1.1 Background of Study
The traditional method of attendance marking in schools and Universities is a tedious and time
consuming task. It often requires to manually call out the names of students or manually sign on sign
sheets, which can take a significant portion of the session. Moreover, this method is prone to errors and
proxy attendance. The purpose of this project is to develop an attendance system based on face
recognition techniques. The system will use individuals' faces for marking attendance. Face recognition
has gained popularity and is widely used in various applications. In this paper, we propose a system that
detects the faces of students from live streaming video by using in a classroom and marks attendance if
the detected face matches the database Singh, S., & Shekhar, S.[1]

Face recognition has emerged as an important biometric feature that is easily acquired and non-
intrusive. Face recognition-based systems offer advantages over other methods as they are relatively
unaffected by facial expressions. These systems can be categorized into verification, which involves
comparing a face image against template face images, and identification, which involves comparing a
query face image against a database of face images. Sharma A., & Aggarwal A. [2] Face recognition
system is a complex image-processing problem in real world applications with complex effects of
illumination, occlusion, and imaging condition on the live images. It is a combination of face detection
and recognition techniques in image analyzes. Detection application is used to find position of the faces
in a given image. Recognition algorithm is used to classify given images with known structured
properties, which are used commonly in most of the computer vision applications.
Computer vision is a field of AI that enables computers and systems to derive meaningful information
from digital images, videos, and other visual inputs and take actions or make recommendations based
on that information. If AI enables computers to think, computer vision helps them to see, observe and
understand. Automated facial recognition was pioneered in the 1960s. Woody Bledsoe, Helen Chan
Wolf, and Charles Bisson worked on using the computer to recognize human faces. Their early facial
recognition project was dubbed "man-machine" because the coordinates of the facial features in a
photograph had to be established by a human before they could be used by the computer for recognition.
Real-time face detection in video footage became possible in 2001 with the Viola–Jones object detection
framework for faces. By 2015 the Viola-Jones algorithm had been implemented using small low power
detector on handheld devices and embedded systems. Bledsoe, W., Wolf, H. C., & Bisson, C .[9]
Therefore, the Viola Jones algorithm has not only broadened the practical application of face recognition
systems but has also been used to support new features in user interfaces and teleconferencing.
In today's world of interne working the need to maintain the security of information or physical property
is becoming both increasingly important and increasingly difficult. From time to time we hear about
prison escapes which is the act of an inmate leaving prison through unofficial or illegal ways, and an
effort is made on the part of authorities to recapture them and return them to their original detainers;
however, it is difficult to recognize them from the crowd, this paves a way to a valuable security
techniques.

1.2 Statement of the problem

Traditional methods of attendance management, such as manual attendance sheets or roll calls, are time-
consuming, error-prone, and lack real-time monitoring capabilities. These methods often lead to
inefficiencies, inaccuracies, and difficulties in managing attendance data. Therefore, there is a need for
an automated attendance system that improves efficiency, accuracy, and data management in attendance
tracking. Attendance management refers to the process of accurately tracking and recording the presence
of individuals in educational institutions. The proposed system aims to utilize face recognition
technology to automate the attendance marking process. capturing and analyzing facial features, the
system can accurately identify and recognize students in real-time. This eliminates the possibility of
proxy attendance and ensures that attendance records are reliable and tamper-proof. The system should
provide a user-friendly interface that allows teachers to easily record attendance, view attendance
reports, and manage attendance data. It should also have the capability to generate comprehensive
reports.
1.3 Objectives

1.3.1 General objectives

To develop a face recognition-based attendance taking system that improves the accuracy,efficiency,
and reliability of attendance tracking in our educational institutions.

1.3.2 Specific objectives

To achieve the overall objective of implementing a face recognition-based attendance system, the
following specific objectives are outlined.
 To design face detection system from the live video footage.
 To design face recognition system using deep neural network algorithm to recognize the wanted
image.
 Implementing face recognition using OpenCV library in Python programming language.
 To interface the program with a video surveillance and perform live face recognition.
 Comparing and analyze the results of the output matching with the input face of the wanted.

1.4 Significance of the project

The significance of implementing a face recognition-based attendance tracking system extends to

various organizations, science, and the country as a whole. For organizations, this system streamlines
administrative tasks by automating the attendance recording process, which reduces the need for manual
intervention and minimizes human error. This automation saves time for both employees and
administrators, allowing them to focus on more critical tasks. Additionally, the system reduces costs
associated with traditional attendance methods, such as the purchase and maintenance of RFID cards or
manual logbooks. By enhancing overall operational efficiency, organizations can ensure accurate and
timely attendance records, leading to better resource management and productivity.

In the scientific community, the project contributes significantly to advancements in computer vision
and machine learning. Developing and deploying a robust face recognition system involves extensive
research in image processing, pattern recognition, and algorithm optimization. This project serves as a
practical application that demonstrates the capabilities and potential of these technologies, inspiring
further research and development. By addressing real-world challenges, it helps bridge the gap between
theoretical research and practical implementation, fostering innovation in the field.

On a broader scale, the adoption of such systems can improve productivity and accountability across
different sectors, including education, corporate, and government institutions. In educational settings,
accurate attendance tracking can lead to better student engagement and performance monitoring. In
workplaces, it ensures employee accountability and can be integrated with other systems for
performance evaluation. For the country as a whole, widespread adoption of advanced technologies like
face recognition can drive technological growth and innovation, positioning the country as a leader in
the tech industry. Moreover, it can lead to the development of new industries and job opportunities,
contributing to economic growth. By promoting the use of cutting-edge technology, the country can
enhance its global competitiveness and ensure a secure, efficient, and forward-looking infrastructure.

1.5 Scope and Contribution of the Project

The system should be built to be used for a prolonged period of time anywhere in the country where
surveillance/CCTV cameras are installed. This project proposed a face recognition system for detecting
and identifying the person who was attended or not. But it can also be used to find and locate missing
persons. The scope of face recognition attendance systems extends across various sectors, including
corporate offices, educational institutions, healthcare facilities, and governmental organizations. .
Healthcare facilities employ these systems to manage staff schedules effectively and maintain
compliance with regulatory requirements. Moreover, government agencies utilize face recognition
technology to enhance security measures and optimize resource allocation. As the technology continues
to evolve, the scope of face recognition attendance systems is expected to expand further, becoming an
indispensable tool for organizations seeking to modernize their workforce management practices.

1.6 Limitation of the project

This project has several limitations. Firstly, the live face recognition system requires network access for
optimal performance, as offline face recognition cannot be conducted. Additionally, the system is unable
to recognize individuals who are wearing eyeglasses or face masks, and it cannot differentiate between
twins. Another limitation is that the project can only identify a person’s face from front views, not side
views, necessitating additional features and algorithms to improve its capabilities
1.7 Thesis Outline
Chapter one: Gives the introduction to the project, statement of the problem, objectives, evaluation
methodology, and the scope of project.
Chapter Two: The literature review which describes the system.
Chapter Three: System design.
Chapter Four: Test results and discussions on the results.
Chapter Five: The overall conclusion of development of the project is enclosed together with
suggestion and recommendation for future work or enhance. This chapter also includes appendices and
the references used.
 CHAPTER TWO

2 Literature Review
Face recognition is one of the most developing research zones because of increasing demands for
security in numerous applications. In this session, we are going to present quick overview of some of the
previous proposed works. Lastly, we will be presenting the advancement of our project compared to
those mention.

2.1 Literature Survey on the Face Detection and Feature Extraction

As the first step in any automatic face recognition systems is the detection of faces in images, there
existed different researches made on this process. Significant advances have been made in recent years
in achieving automatic face detection under various conditions. Compared to feature-based methods and
template-matching methods, appearance or image-based methods that train machine systems on large
numbers of samples have achieved the best results. This may not be surprising since face objects are
complicated, very similar to each other, and different from non-face objects.
More recently, detection of faces under rotation in depth has been studied. One approach is based on
training on multiple view samples Compared to invariant-feature-based methods, multi view-based
methods of face detection and recognition seem to be able to achieve better results when the angle of
out-of-plane rotation is large (35◦).
Another recent advance in this process is the component-based detection/recognition and 3D morphable
models. The basic idea of component-based methods is to decompose a face into a set of facial
components such as mouth and eyes that are interconnected by a flexible geometrical model. The
motivation for using components is that changes in head pose mainly lead to changes in the positions of
facial components which could be accounted for by the flexibility of the geometric model. However, a
major drawback of the system is that it needs a large number of training images taken from different
viewpoints and under different lighting conditions.

2.2 Methodology

In this section, we outline the methodology that we will employ to achieve our project objectives of
improving face recognition for attendance systems. The methodology consists of several key steps:
Face Detection and Recognition Library Selection
in order to perform face recognition with python and opencv we need to install two additional libraries
such as dilb maintained by davis king ,contains our implementation of deep metric learning which is
used to construct our face embedding’s used for the actual recognition process, Face recognition

Webcam Integration
An integrated webcam, is a small camera module that comes pre-installed in devices such as laptops,
tablets, and smartphones. It eliminates the need for an external camera, allowing users to capture video
and still images, engage in video calls, take pictures, and record videos directly without additional
equipment. It can try reinstalling or updating the driver to get it working again. Additionally, on
Windows 11, it can manage connected cameras through the Camera settings Configure the webcam to
capture real-time video frames. Establish the necessary connections with the webcam using Python
libraries like Open CV or Pygame.
Real-Time Face Detection and Recognition
Real-time face detection and recognition involve identifying and locating human faces within digital
images or videos. There are some key points such as Face Detection is a computer vision task that
automatically identifies and locates human faces in images or videos, Real-Time Aspect Modern face
detection algorithms can perform this task in real time, making it suitable for applications like video chat
or games that respond to facial expressions.
Real-time Database Integration
Real-time database integration is a crucial aspect of modern software development. Real-time database
integration involves connecting your application to a database that provides real-time updates. Real-time
databases allow data synchronization across clients and servers instantly, ensuring consistent and up-to-
date information.it Use Collaborative Apps Real-time databases are ideal for collaborative tools like chat
applications, collaborative document editing, and project management tools, Live Dashboards They
enable real-time visualization of data, such as live sales dashboards or monitoring systems, Design an
appropriate database structure within the Firebase Real-time Database to store student information,
attendance records, timestamps, and total attendance
Attendance Recording and Database Management
Implement the necessary functions and procedures to record attendance by linking the face recognition
results with the Firebase Realtime Database.
Utilize the Firebase SDKs and libraries for Python to store student information, attendance records,
timestamps, and total attendance within the Firebase Realtime Database.

User Interface A Facial Recognition Attendance Monitoring System is designed to address the
limitations of conventional attendance tracking methods. It leverages cutting-edge advancements in
computer vision and machine learning to provide a robust and reliable mechanism for recording
attendance. Here are some key features of such a system Develop a user-friendly interface for
instructors/administrators. Display real-time attendance updates.

2.3 Evaluation

2.3.1 Accuracy
measures the proportion of correctly identified faces out of the total faces the accuracy of a face
recognition-based attendance system depends on several factors, including the quality of the dataset, the
chosen face recognition algorithm, and the preprocessing techniques used .there are Data Collection and
Preprocessing, Face Recognition Algorithms, Real-Time Recognition.

Formulae accuracy = (true positive +true negative)/total sample

2.3.2 Precision

Precision quantifies the ratio of true positive predictions (correctly identified faces) to the total positive
predictions (both true positives and false positives).

Formula: Precision = True Positives}\text {True Positives + False Positives}}

Recall (Sensitivity): (also known as sensitivity or true positive rate) represents the ratio of true positive
predictions to the total actual positive instances (both true positives and false negatives).

Formula: Recall =True Positives\ {True Positives + False Negatives}

 CHAPTER THREE

3 System Design and Analysis

3.1 Introduction
deep neural networks and convolutional neural networks (CNNs) have become the most commonly used
method for feature representation and have achieved good results in face recognition problems. Face
recognition can be divided into two categories: face verification, where two faces are presented and the
system needs to verify whether these two faces belong to the same person, and face identification, where
a face image is presented with an unknown identity and the system needs to determine this identity.

3.2 DESIGN OF A FACE RECOGNITION SYSTEM

A throughout survey has revealed that various methods and combination of these methods
can be applied in development of a new face recognition system. Among the many
possible approaches, we have decided to use a combination of knowledge-based methods
for face detection part and neural network approach for face recognition part.

Figure 3.1 neural network approach face recognition.

3.3 Input Part

Input part is prerequisite for face recognition system. Image acquisition operation is
performed in this part. Live captured images are converted to digital data for performing
image-processing computations. These captured images are sent to face detection
algorithm.

3.4 Face Detection Part

Face detection performs locating and extracting face i m a g e operations for face recognition
system.
For this reason, skin segmentation is applied as a first step of detection part. RGB color space
is used to describe skin like color.

Fig 3.2 Algorithm of face detection part Chiranjit Dutta, Ranjeet Singh.[6].

White balance of images differs due to change in lighting conditions of the environment while
acquiring image. This situation creates non-skin objects that belong to skin objects.
Therefore, white balance of the acquired image should be corrected before segmenting.

Fig 3.3 image in different white balance.

Final verification of candidate and face image extraction, facial feature extraction process
is applied. Facial feature is one of the most significant features of face. Facial features
are eyebrows, eyes, mouth, nose, nose tip, cheek, etc. The property is used to extract the eyes
and mouth which, two eyes and mouth generate isosceles triangle, and distance between
eye to eye and midp oint of eyes distance to mouth is equal.

Figure 3.4 Facial Feature Extractions (Left) and Face Image (Right).

3.5 Face Recognition Part

Modified face image which is obtained in the Face recognition system, should to be classified to identify
the person in the database. Face recognition part is composed of preprocessing face image,
vectorizing image matrix, database generation, and then classification. The classification is achieved
by using Feed Forward Neural Network (FFNN) Turk, M., & Puntland, A.[7].

Fig 3.5 face recognition part Jalled, F.[15].

Face recognition part algorithm In classifier, FFNN is the simplest structure in the neural network.
This type of network structure is generally used for pattern recognition applications. System network
properties are: input layer has 900 inputs, hidden layer has 41 neurons and output layer has 26
neurons. Output layer has 26 neuron since the number of people in database is 26 Lawrence,

Feedback Mechanism This provides feedback to users or subjects being recognized. For instance, it
may display a confirmation message when a face is successfully recognized or alert the user if there's a
failure or discrepancy.

Security Measures: These ensure the integrity and security of the system. This may include encryption
of communication channels, access control mechanisms to prevent unauthorized usage and measures to
protect.
 .

Fig 3.6 Block diagram of face recognition attendance system

3.6 Working principle

A face recognition attendance system is a type of biometric technology that uses artificial
intelligence to automatically identify and verify individuals based on their facial characteristics.
These systems are commonly used to track attendance in workplaces, stores, and other
organizations.
 Fig 3.7 Biometric face recognition how it work

3.7 Simulation design and analysis

Designing and analyzing a face recognition attendance system involves several steps, including
system architecture design, algorithm selection, simulation setup, and performance evaluation.
Here's a general outline of how you might approach this.

3.8 System Architecture Design

Define the overall architecture of the system, including the components mentioned earlier
camera module, image processing, face recognition algorithm, database, etc. Determine the
interactions between these components and how data flows through the system.

3.9 Algorithm Selection

Choose appropriate face detection and recognition algorithms based on your requirements and
constraints. This might involve researching and experimenting with various techniques, such as
Haar cascades, HOG (Histogram of Oriented Gradients), or deep learning-based methods like
Convolutional Neural Networks (CNNs). Consider factors such as accuracy, speed, resource
requirements, and robustness to variations in lighting, pose, and facial expressions.
3.10 Simulation Setup
Set up a simulation environment using software tools such as OpenCV, Dlib, for implementing
the face recognition algorithms and simulating the system's behavior. Generate synthetic or real-
world datasets for testing and evaluation. These datasets should include images or video
sequences containing faces captured under different conditions (e.g., varying lighting, angles,
and facial expressions).

3.10.1 Implementation
Implement the components of the face recognition attendance system within the simulation
environment. This includes modules for image capture, face detection, face recognition, database
management, and attendance logging. Integrate the selected face detection and recognition
algorithms into the system and fine-tune their parameters as needed.

3.10.2 Performance Evaluation

Define metrics for evaluating the performance of the system, such as accuracy, precision, recall,
processing speed, and resource utilization.Conduct experiments to assess the system's
performance under various conditions, including different lighting conditions, camera angles,
and numbers of individuals in the scene. Analyze the results of the experiments to identify
strengths, weaknesses, and areas for improvement in the system.

3.10.3 Validation and Verification

Validate the performance of the simulated system against ground truth data or benchmarks to
ensure its accuracy and reliability. Verify that the system meets the specified requirements and
objectives, including accuracy targets and operational constraints.

3.10.4 Optimization and Iteration

Based on the performance evaluation results, optimize the system design, algorithms, and
parameters to improve its overall effectiveness. Iterate on the design and implementation as
necessary, incorporating feedback and lessons learned from the evaluation process. By following
these steps, you can design, simulate, and analyze a face recognition attendance system to
understand its capabilities, limitations, and potential for real-world deployment.

3.11 Software and hardware component used

3.11.1 Hardware component
Webcam is a video camera that feeds or streams an image or video in real time to or through
a computer network, such as the Internet. Webcams are typically small cameras that sit on a
desk, attach to a user's monitor, or are built into the hardware . A webcam is also defined a simple
video camera that, when attached to a computer, lets you capture still images and video. It's
commonly used for video conferencing with software like Zoom, Google Meet, Skype, and other
apps .

3.11.2 Software requirement

This project is done in window 11 platform using python 3.7 software and to accomplish this
project we use different libraries including

OpenCV is a powerful open-source framework for computer vision tasks. It provides over 2500
algorithms for real-time applications and is widely used in face recognition systems this step
locates the position of human faces within the image. OpenCV is an image and video processing
library and is used for image and video analysis. The dlib library, maintained by Davis King,
contains our implementation of “deep metric learning” which is used to construct our face
embedding used for the actual recognition process.

Dlib is a popular open-source library for machine learning and computer vision tasks. It provides
tools for face detection, face recognition, and other related tasks. To create a face recognition-
based attendance system using Dlib.

Face-recognition is a technology that identifies and verifies individuals based on their facial
features. It analyzes patterns in the face, such as the distance between eyes, nose shape, and
jawline, to create a unique representation (often called a face template).

CMake is an open-source, cross-platform family of tools designed to build, test and package
software. CMake is used to control the software compilation process using simple platform and
compiler independent configuration files, and generate native make files and workspaces that can
be used in the compiler environment of your choice,

CV Zone provides self-paced courses covering topics like self-driving, object detection,
augmented reality, robotics.

CMake is an open-source, cross-platform family of tools designed to build, test and package
software. CMake is used to control the software compilation process using simple platform and
compiler independent configuration files, and generate native make files and workspaces that can
be used in the compiler environment of your choice.

3.12 System analysis

System analysis of a face recognition attendance tracking system involves examining its various
components, processes, interactions, and performance characteristics to ensure its effectiveness,
reliability, and efficiency. Here's a systematic approach to conducting system analysis for such a
system:

Requirements Analysis Is functional and non-functional requirements of the system, including

features such as face detection, recognition accuracy, processing speed, scalability, security, and
user interface. Gather requirements from stakeholders, including end-users, administrators, and
IT personnel, to ensure that the system meets their needs and expectations.

Use Case Modeling Identify the different actors (e.g., users, administrators) and use cases (e.g.,
enroll new user, mark attendance) of the system.Create use case diagrams to visualize the
interactions between actors and the system and to capture the functional requirements from a
user's perspective.
 Fig 3.8 use case model of face recognition attendance system

System Architecture Analysis Analyze the architecture of the system to understand its overall
structure, components, and interactions. Identify the key subsystems/modules, such as image
capture, face detection, recognition engine, database management, and attendance logging.
Evaluate the design choices regarding the distribution of responsibilities, communication
protocols, and data flow between subsystems.
 Fig 3.9 System architecture analysis

Data Flow Analysis Model the flow of data through the system, from the initial capture of
images to the final attendance record. Identify data sources, processing steps, transformations,
storage mechanisms, and interfaces between components. Analyze potential bottle necks, data
dependencies, and points of failure in the dataflow.

The system operates in a series of steps to the following methods:

User Registration →Real-Time Face Detection → Face Recognition → Attendance Marking →
Report Generation.

Below is a simplified flow of the system's operation

1. User Registration

2. Face Detection (from Camera

 3. Face Recognition (Compare with Database)

4. Match Found? --> Yes --> Mark Attendance (Log Data)

No --> Flag as Unknown

5. Send Notification (Real-time Update

6. Generate Attendance Reports (Daily/Weekly)

7. Data Security and Backup (Encrypted & Stored)

8. Manual Verification (if required)

9. End of Session (Store Final Logs)

Flow chart of Face recognition system

Figure flow chart

Performance Analysis Define performance metrics such as recognition accuracy, processing
time, throughput, and resource utilization. Conduct performance testing under various scenarios,
including different lighting conditions, camera angles, and numbers of individuals. Measure the
system's response time, scalability, and robustness to variations in input data and work load.

Security Analysis Evaluate the security measures implemented in the system to protect sensitive
data, such as biometric templates and attendance records. Assess the effectiveness of
authentication mechanisms, encryption techniques, access controls, and audit trails in preventing
unauthorized access and ensuring data integrity.

Usability Analysis Assess the usability of the system from the perspective of end-users and
administrators. Conduct usability testing and gather feedback on aspects such as user interface
design, ease of use, intuitiveness, and learn ability .Identify any usability issues or pain points
and propose improvements to enhance user experience.

Reliability and Availability Analysis Evaluate the reliability and availability of the system to
ensure continuous operation and minimal downtime. Analyze factors such as fault tolerance,
error handling, system recovery and backup strategies to mitigate potential failures and
disruptions.

Cost-Benefit Analysis Perform a cost-benefit analysis to assess the economic feasibility of the
system, including upfront development costs, ongoing maintenance expenses, and potential
benefits such as labor savings and improved accuracy. Compare the costs and benefits of the
face recognition attendance system with alternative solutions or manual processes to determine
its value reposition.
Risk Analysis Identify potential risks and uncertainties associated with the system such as
technical challenges regulatory compliance, privacy concerns, and organizational resistance.
Assess the likelihood and impact of each risk and develop mitigation strategies to minimize their
effects on the project's success. By conducting thorough system analysis, you can gain insights
into the strengths, weaknesses, opportunities, and threats of the face recognition attendance
tracking system and make informed decisions to optimize its design, implementation, and
deployment.
3.13 Challenge faced during software and hardware implementation
Implementing a face recognition attendance system faces several challenges. One challenge is
the variability in real-world scenarios, including pose, lighting, and expression variations, which
can affect the accuracy of face recognition systems . Another challenge is the need for advanced
algorithms and machine learning techniques to analyze and compare facial features, in order to
overcome these variations and improve system performance . Additionally, the system must be
able to handle multiple users in one frame and detect faces accurately in low lighting conditions .
Furthermore, the system should be contactless, especially in the current pandemic era, to ensure
the safety and health of users. One of the main challenges of face detection and recognition is
the high variability of human faces in terms of shape, size, pose, expression, illumination,
occlusion, and makeup. These factors can make it difficult for the algorithms to generalize and
cope with different scenarios and condition.

3.14 Measure to overcome the problems

Robust Algorithm Selection: Choose a robust face recognition algorithm that performs well
under various conditions such as varying lighting, facial expressions, and occlusions. Deep
learning-based algorithms like Convolutional Neural Networks (CNNs) often perform well.

Quality Data Collection: Gather a diverse dataset for training your face recognition model.
Ensure it includes various demographics, facial expressions, accessories (glasses, hats), and
lighting conditions. High-quality data is crucial for accurate recognition.

Preprocessing Techniques: Apply preprocessing techniques to enhance image quality before

feeding it into the face recognition model. Techniques like normalization, histogram
equalization, and noise reduction can improve recognition accuracy.

Hardware Compatibility: Ensure that the hardware components (e.g., cameras, processors) are
capable of handling the computational requirements of the face recognition algorithm. Opt for
hardware acceleration techniques like GPUs or specialized chips if needed.
Real-time Processing: Aim for real-time processing to achieve seamless attendance tracking.
Optimize algorithms and hardware to minimize latency and processing time.

Privacy and Security: Implement robust privacy and security measures to protect the data
collected by the face recognition system. Ensure compliance with relevant regulations such as
GDPR or CCPA.

Testing and Validation: Conduct thorough testing and validation of the system under various
scenarios to identify and address any issues. Use techniques like cross-validation and
performance metrics evaluation to assess the system's accuracy and reliability.

Continuous Improvement: Implement mechanisms for continuous improvement of the face

recognition system. This includes collecting feedback from users, monitoring system
performance, and updating algorithms or hardware as needed.

User Interface Design: Design an intuitive user interface for easy interaction with the
attendance system. Ensure that it provides clear feedback to users and supports functionalities
like enrollment, authentication, and reporting.

Training and Support: Provide adequate training and support to users and administrators of the
face recognition attendance system. This includes training on system usage, troubleshooting
common issues, and handling exceptions.

3.15 Purpose of the system design

The software can be used for security purposes in organizations and in secured zones. The
software stores the faces that are detected and automatically marks attendance. The system is
convenient and secure for the users. It saves their time and efforts.
By identifying individuals based on unique facial features, the system significantly reduces the
risk of errors and ensures reliable attendance tracking. Time Efficiency and Productivity: The
real-time processing capability of the face recognition system speeds up the attendance tracking
process.
Algorithm Selection
Software: Choose a face recognition algorithm that suits the requirements of your application,
such as Convolutional Neural Networks (CNNs) for deep learning-based approaches or Eigen
faces for traditional methods. Hardware select hardware components capable of efficiently
running the chosen algorithms. Consider factors like computational power, memory bandwidth,
and parallel processing capabilities.
Hardware Acceleration
Software: Implement optimizations in the software to leverage hardware acceleration where
possible. This may involve using libraries like OpenCV or TensorFlow that support hardware
acceleration through frameworks like CUDA or OpenCL.
Hardware: Use specialized hardware accelerators such as Graphics Processing Units (GPUs),
Field-Programmable Gate Arrays (FPGAs), or dedicated AI chips (e.g., TPUs, NPUs) to
accelerate the execution of computationally intensive tasks like convolution operations in neural
networks.
Data Processing:
Software: Develop software components for data preprocessing, feature extraction, and
classification using the chosen face recognition algorithm. Optimize these components for
efficient execution on the target hardware.
Hardware: Ensure that the hardware components have sufficient processing power and memory
bandwidth to handle the data processing requirements of the software algorithms. Consider
parallel processing capabilities to speed up computations.
Input and Output Interfaces:
Software: Implement input interfaces for capturing images or video frames from cameras, as well
as output interfaces for displaying recognition results or feedback to users.
Hardware: Choose compatible hardware components for input and output interfaces, such as
cameras with appropriate resolution and frame rates, displays or monitors for visual feedback,
and input devices for user interaction.

Integration and System Design:

Software: Design the software architecture to efficiently integrate with the selected hardware
components. Ensure that communication interfaces (e.g., drivers, APIs) are available to interact
with hardware devices and perform necessary operations.
Hardware: Plan the integration of hardware components into the overall system architecture,
considering factors like power consumption, thermal management, and physical connectivity.
Testing and Optimization:
Software: Conduct thorough testing of the software on the target hardware platform to identify
performance bottlenecks, optimize algorithms, and fine-tune parameters for optimal
performance.
Hardware: Benchmark the hardware components to assess their performance under different
workloads and configurations. Adjust hardware settings or configurations as needed to achieve
optimal performance for the face recognition system.
 CONCLUSION
Face recognition systems are part of facial image processing applications and their significance
as a research area are increasing recently. Implementations of system are crime prevention, video
surveillance, person verification, and similar security activities. The face recognition system
implementation can be part of Universities. Face Recognition Based Attendance System has
been envisioned for the purpose of reducing the errors that occur in the traditional (manual)
attendance taking system. The aim is to automate and make a system that is useful to the
organization such as an institute. The efficient and accurate method of attendance in the office
environment that can replace the old manual methods. This method is secure enough, reliable and
available for use. Proposed algorithm is capable of detect multiple faces, and performance of
system has acceptable good results
 REFERENCES
[1]. A brief history of Facial Recognition, NEC, New Zealand,26 May 2020.[Online]. Available:
https://www.nec.co.nz/market-leadership/publications-media/a-brief-history-of-
facialrecognition/
[2]. Face detection,TechTarget Network, Corinne Bernstein, Feb, 2020.[Online]. Available:
https://searchenterpriseai.techtarget.com/definition/face-detection
[3]. Paul Viola and Michael Jones, Rapid Object Detection using a Boosted Cascade of Simple
Features. Accepted Conference on Computer Vision and Pattern Re cognition, 2001.
[4]. Face Detection with Haar Cascade,Towards Data Science-727f68dafd08,Girija Shankar
Behera, India, Dec 24, 2020.[Online]. Available:https://towardsdatascience.com/face-
detectionwith-haar-cascade-727f68dafd08
[5]. Face Recognition: Understanding LBPH Algorithm,Towards Data
Science90ec258c3d6b,Kelvin Salton do Prado, Nov 11, 2017.[Online]. Available
:https://towardsdatascience.com/face-recognition-how-lbph-works-90ec258c3d6b