VISVESVARAYA TECHNOLOGICAL UNIVERSITY
“JNANASANGAMA”, BELAGAVI-590018
K R PETE KRISHNA GOVERNMENT ENGINEERING COLLEGE, KR PET-571426
2023-2024
Department of Computer Science & Engineering
Project Presentation on
“FACE RECOGNITION BASED ATTENDENCE SYSTEM”
Presented By
DEEPIKA M V 4GK21CS015
LOKESH V N 4GK21CS019
BASAVARAJ BELUR 4GK22CS400
Under the Guidance of Head of the Department
Dr. Shivashankar Dr. Naggapa U Bhajanthri
Assistant Professor Head of the Department
Dept. of CSE Dept. of
CSE
CONTENTS
o Abstract
o Introduction
o Literature Survey
o Existing System
o Problem Statement
o Objectives
o Architecture
o Advantages
o Future Scope
o Conclusion
ABSTRACT
The “Face Recognition Based Attendance System” aims to automate the
attendance tracking process in educational institutions and workplaces by
leveraging advanced facial recognition technology.
Traditional attendance methods are often time-consuming and prone to errors.
Our system utilizes machine learning algorithms to accurately identify
individuals from live camera feeds, ensuring a quick and reliable way to record
attendance,
The project involves the integration of a webcam with a backend system that
processes facial images to match them with stored profiles.
It will feature user-friendly interfaces for both students and administrators,
allowing easy monitoring of attendance records.
INTRODUCTION
A “Face Recognition based Attendance System” is an innovative approach to
automating attendance tracking using facial recognition technology.
This system leverages machine learning and computer vision to identify and verify
individuals based on their facial features, offering a seamless, contactless, and efficient
solution for attendance management.
It eliminates the need for manual attendance recording methods like roll calls or sign-
ins, reducing human error and saving time.
The system captures an image of the individual, processes it using facial recognition
algorithms, and matches it with stored data to confirm identity. Once recognized,
attendance is automatically logged into a database.
This technology finds applications in educational institutions, workplaces, and events,
where accurate and real-time attendance tracking is critical.
LITRATURE SURVEY
Ref Year Title of the Methodolo Remarks
No.
Paper gies/
Technologi
es Used
[1 202 Design and Haar -Like Pros :Achieves 86% accuracy in face
Implementation of Features recognition, making it effective for
] 0 Student Attendance Methods
System Based on
attendance tracking.
Face Recognition by Cons : Accuracy is affected by poor or
Haar-Like Features inconsistent lighting conditions.
Methods.
[2 202 Local Binary Local Binary Pros : Streamlines attendance-taking,
Pattern Histogram Pattern saving time during class sessions.
] 1 for Face methods.
Recognition in
Cons : Effective recognition depends on
Student Attendance optimal camera placement.
System .
[3 202 Student Attendance Face Pros :Reduces attendance data processing
using Face recognition time significantly, from 6-8 days to 3-4
] 2 Recognition technology
Technology through
days .
Computer Cons : Accuracy is impacted by poor
Vision. image quality or lighting conditions.
HAAR CASCADE ALGORITHM
The Haar Cascade algorithm is a set of classifiers used for object detection.
The Haar cascade algorithm is a machine learning technique that detects objects in
images, regardless of their scale or location.
The images which we would like to be classified by the classifier are known as
positive images and the images we would not want our classifier to classify are
known as negative images.
The algorithm uses a cascading window to compute features in each window and
classify whether it contains an object. It creates a "strong classifier" by combining
"weak classifiers".
LBPH ALGORITHM
A "Local Binary Pattern Histogram" (LBPH) algorithm is a technique used primarily
for face recognition, where it analyzes the texture of an image by comparing each
pixel to its surrounding neighbors, creating a binary pattern that is then converted
into a histogram, effectively capturing the local features of the image, making it
robust to changes in lighting and minor variations in facial expressions
EXSISTING SYSTEM
1.Traditional Attendance Systems: Manual attendance-taking methods, such as
paper-based sign-in sheets or RFID-based systems, have limitations in terms of
accuracy, efficiency, and scalability.
2. Biometric-based Attendance Systems: Fingerprint, iris, and facial recognition
technologies have been explored for attendance tracking. Facial recognition has
gained popularity due to its non-invasive nature.
LIMITATIONS OF EXISTING SYSTEM
1. Lighting Dependency: The system requires good lighting conditions for the camera to
capture clear images. Poor lighting can affect the accuracy of facial detection and
recognition.
2. Scalability: The current system is designed to handle a limited number of users (up to
20)in the prototype. While it is scalable, the system in its current form is not optimized for
large classes without additional modifications.
3. Identical Twins Issue: The system struggles with differentiating between identical
twins . Since twins may have highly similar facial features, the system may mistakenly
identify one as the other.
4. Appearance Changes: If a student changes their appearance significantly (e.g.,
hairstyle , facial hair, or makeup), the system may fail to recognize them and mark their
attendance incorrectly
PROBLEM STATEMENT
“Create an automated, face recognition-based attendance system to
improve accuracy and efficiency, eliminating the common errors and delays
associated with manual attendance tracking.“
"Manual attendance tracking in educational institutions is time-consuming,
prone to errors, and vulnerable to proxy attendance. Existing systems lack automation,
leading to inefficiencies and security concerns. This project aims to develop a reliable and
efficient Face Recognition-Based Attendance System to overcome these challenges."
PROPOSED SYSTEM
The proposed Face Recognition Based Attendance System leverages advanced image
processing techniques, specifically the Haar-like features method and Local Binary
Pattern Histogram (LBPH), to provide an efficient and accurate solution for attendance
management.
The Haar-like features method is a popular technique used for object detection,
particularly in face detection. This method involves the use of simple rectangular features
to detect the presence of a face in an image. These features are calculated by taking the
difference between the sum of pixel intensities in adjacent rectangular regions.
The advantage of using Haar-like features is their ability to capture the essential
characteristics of a face, such as the eyes, nose, and mouth, with minimal computational
complexity.
OBJECTIVES
Increase efficiency: By automating attendance, face recognition systems can save time
and reduce errors.
Improve security: By identifying authorized personnel, face recognition systems can
prevent unauthorized people from entering the company.
Reduce costs: Face recognition systems can eliminate the need for additional hardware
and maintenance costs.
Manage attendance remotely: Face recognition systems can be used to track attendance
remotely.
Track attendance offline: Face recognition systems can track attendance offline.
Manage attendance across multiple locations: GPS-based face recognition systems can
be used to monitor attendance across multiple locations .
ARCHITECTURE
ADVANTAGES
Accuracy:
•Reduces errors associated with manual attendance systems.
•Ensures that only the registered individuals are marked present, preventing
fraudulent entries.
Efficiency:
•Speeds up the attendance process, saving time for both students and staff.
•Reduces administrative workload, allowing staff to focus on more important
tasks.
Convenience:
•Eliminates the need for physical attendance sheets or cards.
•Allows for seamless integration with existing database systems for real-time
record updates.
Scalability:
•Easily adaptable to various environments, including schools, universities,
offices, and events.
Eco-Friendly:
•Reduces paper usage by digitizing the attendance process.
•Promotes a more sustainable approach by minimizing physical resources.
ISSUES
Privacy concerns: Collecting and storing facial data raises significant privacy
concerns, as this information can be easily misused or leaked.
Accuracy issues: Facial recognition systems can struggle with variations in
lighting, facial expressions, hairstyles, or even slight changes in appearance, leading
to false positives or negatives.
Bias in algorithms: Studies have shown that facial recognition algorithms can
exhibit bias towards certain demographics, leading to unfair identification errors.
Outputs
FUTURE SCOPE
1.Improving Robustness: Developing algorithms resilient to variations in
lighting, pose, and expression.
2. Real-time Processing: Optimizing algorithms for real-time face recognition.
3. Security Enhancements: Implementing secure data storage and transmission
protocols.
CONCLUSION
"The Face Recognition-Based Attendance System successfully automates
attendance tracking with accuracy and efficiency. This innovative solution
reduces manual errors, enhances security, and streamlines institutional
operations. Its scalability and versatility make it an ideal solution for
educational institutions, corporate sectors, and government organizations. This
project demonstrates the potential of facial recognition technology in
transforming attendance management”
REFERENCE
1.Dhanush Gowda H.L, K Vishal, Keertiraj B. R, Neha Kumari Dubey,Pooja M. R.International
Journal of Engineering Research & Technology (IJERT), International Journal for Research in
Applied Science & Engineering Technology (IJRASET) in 2019.
2.Soundarya S1,Ashwini P2,Rucha W3,Gaurav K4,MS.Savitri Patil Department Of
Information Technology,G.H. Raisoni(GHRCEM) Pune,India, Volume 9, Issue 11 November
2021 | ISSN: 2320-2882
3. Evta Indra, Muhammad Yasir, Andrian, Delima Sitanggang, Oloan Sihombing, Saut Parsoran
Tamba, Elviana Sagala,[August 15,2022]
THANK YOU!!
