CAPSTONE PROJECT

FACIAL RECOGNITION ANSHIKA SINGH

ATTENDANCE
MONITORING SYSTEM here please add
a picture of yours
PRESENTED BY

STUDENT NAME: DIGVIJOY RANJAN

COLLEGE NAME: RAJ KUMAR GOEL INSTITUTE OF

TECHNOLOGY AND MANAGEMENT

DEPARTMENT: BACHELOR OF TECHNOLOGY IN COMPUTER

SCIENCE AND ENGINEERING

EMAIL ID: [email protected]

AICTE STUDENT ID: STU61d1487daaf121641105533

 OUTLINE
• Problem Statement
• Proposed System/Solution
• System Development Approach
• Algorithm & Deployment
• Result Conclusion
• Future Scope
• References
 PROBLEM STATEMENT
Manual attendance systems are slow, error-prone, and vulnerable to proxy
entries. They lack transparency, efficiency, and real-time tracking. With
rising demand for contactless and accurate solutions, there’s a need for an
intelligent system that ensures secure and automated attendance
monitoring.
 PROPOSED SOLUTION
The proposed system uses facial recognition technology to automate and
secure the attendance process. It leverages computer vision and machine
learning to detect and recognize faces in real time. Once a face is identified,
the system marks attendance instantly with timestamped records, eliminating
manual errors and proxy entries. The solution includes:
•Data Collection: Captures facial images via webcam in real-time.
•Face Detection & Recognition: Uses OpenCV and deep learning models to
identify unique facial features.
•Database Logging: Stores attendance data securely with time and date.
•User Interface: A simple GUI for administrators to register faces and
monitor logs.
•Scalability: Designed to handle multiple users and extended deployment in
institutions or organizations.
 SYSTEM APPROACH
The system follows a modular approach combining real-time facial
recognition with automated attendance logging. It integrates computer
vision, machine learning, and GUI development to ensure a robust and user-
friendly solution.
•System Requirements: OS: Windows/Mac, Python 3.10.11, Webcam
(built-in or external)
•Libraries/Technologies Used: OpenCV – for face detection and
recognition, NumPy, Pandas – for data manipulation, Tkinter – for GUI
development, Pillow – for image handling, Datetime – for timestamping
attendance, CSV – for data logging and export
o Tk-tools, opencv-contrib-python, datetime, pytest-shutil, python-csv,
numpy, pillow, pandas, times.
•Architecture: Face capture and encoding, Real-time face recognition,
Attendance logging to CSV, GUI for user interaction.
 ALGORITHM & DEPLOYMENT
Algorithm Used: The system uses the HOG (Histogram of Oriented
Gradients) and CNN (Convolutional Neural Network) models via the dlib and
face_recognition libraries for robust face detection and recognition.
•HOG is lightweight and effective for real-time detection.
•CNN improves accuracy in complex lighting and pose variations.
Data Input: Real-time webcam feed, Pre-encoded facial data (for known
individuals)
Workflow: 1. Capture real-time image via webcam, 2. Detect face using
HOG/CNN, 3. Encode facial features into a numerical array, 4. Match with
stored encodings, 5. Mark attendance with timestamp (if matched).
Deployment: Built as a Python-based standalone desktop application,
Uses Tkinter GUI for user interaction, Attendance records saved in CSV
format for analysis, Can be scaled for institutions with multiple users and
systems
 RESULT
The facial recognition attendance system was successfully implemented and
tested. It accurately detects and recognizes pre-registered faces and marks
attendance with precise timestamps. Key Outcomes:
•Accuracy: Achieved over 95% recognition accuracy in well-lit conditions.
•Speed: Face recognition and attendance logging completed in under 2 seconds
per user.
•Security: Prevents proxy attendance through unique facial feature encoding.
•Usability: Simple and intuitive interface for admin use and student registration.
Visual Output: Screenshots of face recognition in action, Sample CSV
attendance log with date & time, GUI interface during real-time recognition.
Note: The system was tested across multiple sessions and environments to
ensure consistency and robustness.
Importing Modules
Functions
Function for Email Send
GUI Interface
Sample CSV attendance log with
date & time
 CONCLUSION
The Facial Recognition Attendance Monitoring System successfully
addresses the limitations of traditional attendance methods by providing a
fast, accurate, and contactless alternative. By leveraging artificial
intelligence and computer vision, the system ensures real-time face
detection and recognition with high accuracy, eliminating issues such as
proxy attendance and manual errors. During implementation, challenges
such as varying lighting conditions and initial setup complexities were
encountered, but were effectively managed. The project highlights the
potential of AI in automating repetitive tasks and improving efficiency in
academic environments. Future enhancements could include cloud-based
data storage, support for mask detection, and the development of mobile
or web-based applications for broader accessibility.
 FUTURE SCOPE
•Integrate cloud storage for centralized, real-time attendance data access

•Implement deep learning models for higher accuracy in varied conditions

•Add mask detection and multi-face recognition for robustness

•Extend to mobile and web platforms for wider accessibility
•Integrate with ERP/LMS systems for seamless academic management
•Use edge computing for faster, offline face recognition
•Enable emotion or attention tracking for advanced classroom analytics
•Expand deployment across multiple institutions or locations
 REFERENCES
•GitHub Repository (Face Recognition Library):
https://github.com/ageitgey/face_recognition
•OpenCV Documentation (Computer Vision Library):https://docs.opencv.org/
•Python Official Documentation (Language
Reference):https://docs.python.org/
•Pandas Documentation (Data Handling and
Analysis):https://pandas.pydata.org/docs/
•Dlib Library (Face Detection & Recognition Toolkit):http://dlib.net/
•NumPy Documentation (Scientific Computing Library):https://numpy.org/doc/
•Tkinter Documentation (Python GUI Library):
https://docs.python.org/3/library/tkinter.html
Thank you