FACIAL RECOGNITION BASED
ATTENDANCE/SECURITY SYSTEM
A Project Presentation
submitted in partial fulfillment of the requirements for the
degree of
BACHELOR OF TECHNOLOGY IN
COMPUTER SCIENCE & ENGINEERING
Presented by: Under the guidance
Karan Soni of:
Dr. Sunita Choudhary
INTRODUCTION
Face recognition has emerged as a
crucial technology in biometric
security and automated
authentication. The Face Recognition
Security/Attendance System aims to
provide a reliable and efficient way
of tracking attendance without
manual intervention. The system
uses deep learning-based face
recognition models to detect,
extract, and compare facial features
for identification. The project also
integrates security measures, live
attendance tracking, and a
structured database for easy
retrieval of records
• The Facial Recognition
Smart Attendance and
Security System is a
biometric-based solution
designed to enhance
security and automate
attendance tracking using
artificial intelligence (AI)
and computer vision.
• This system leverages
deep learning-based facial
recognition models to
detect, extract, and verify
facial features for
accurate identification
• With increasing advancements
in artificial intelligence,
machine learning, and
biometrics, traditional
attendance systems such as
manual registers, RFID cards,
and fingerprint scanning are
becoming obsolete due to
inefficiencies, security
loopholes, and hygiene
concerns.
• A facial recognition-based
system ensures a seamless,
contactless, and fraud-proof
mechanism for attendance
• This system finds
applications in various
sectors, including
educational institutions,
corporate offices,
government organizations,
and security-sensitive
environments.
• It eliminates the risks of
proxy attendance, identity
fraud, and unauthorized
access while providing real-
time monitoring and
automated logging of
attendance records.
• In traditional attendance
systems, manual marking of
attendance is time-
consuming and prone to
errors.
• This project proposes an
automated smart
attendance system using
facial recognition technology
to improve accuracy and
efficiency.
• The system captures
images, detects faces,
recognizes individuals, and
logs attendance in a
System
Overview
• Register New Face Module – Allows users to register by capturing multiple face
images and storing them in the database.
System
Overview
• When a new user (student, employee, or authorized person) needs to be added to
the system, their face is registered.
• The system captures multiple images from different angles to ensure robust
recognition
• Each captured face is processed using OpenCV and Dlib to extract facial landmarks
• The extracted features are converted into a 128D feature vector using a deep
learning model
• The vector is stored in the database, associating it with a unique ID, name, and
other details.
• The captured images are not stored permanently to maintain privacy; only 128D feature
vectors are saved.
• The system continuously scans the
environment using a webcam or CCTV
camera.
• When a person is detected, their face is
compared with stored feature vectors.
• If the face is recognized, the system
updates the Attendance Table in real time.
• If a face is not recognized, the system either
prompts for registration or denies access.
• Live attendance marking eliminates manual
input and prevents proxy attendance
(fraud).
Live Attendance
Logging
• The database is updated in real-time whenever a face is
detected.
• The Attendance Table logs the entry and exit times of
individuals.
• A Tracker Sheet (CSV/Excel) can be generated for further
analysis.
The system uses SQLite
as the database
management system
due to its lightweight
nature and easy
integration with Python.
The database consists of
multiple tables to store
and manage different
types of data.
Integration with RFID for Multi-Factor
Authentication
• Although face recognition is a
secure method for attendance
marking, multi-factor
authentication (MFA) can
enhance security even further.
• RFID (Radio-Frequency
Identification) can be integrated
as an additional verification step.
Integration with RFID for Multi-Factor
Authentication
• The user scans their RFID card, and
their face is simultaneously verified.
• If both match, attendance is marked;
if there’s a discrepancy, the system
flags the attempt.
• Reduces chances of spoofing (fake
face attempts)
THANK YOU
