JSPM UNIVERSITY, PUNE

FACULTY OF SCIENCE AND TECHNOLOGY

SCHOOL OF COMPUTATIONAL SCIENCE

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Course Name: IT Infrastructure Landscape Overview

Lab Course Code: 230GCSB68 03

Program: SY B.Tech(CSE) AY 2023-24

Group No./ID-1. Smart Attendance System Using Face Recognition

Name of Team Leader with PRN: ANIKET CHOUDHARI(22358010061)

Name of Team Members with PRN: I. GAURAV BORKAR(22358010046)

2 .SUMIT ANDHARE(22358010006)

3 .LAXMAN CHOUDHARI(22358010052)

4.ATHARV BEDARKAR(22358010019)
1 . Project Title :
Smart Attendance System Using Face Recognition

2. Introduction :
Attendance tracking is a critical component in educational and workplace environments.
Traditional methods such as manual roll-calling or ID card scanning are time-consuming and
prone to errors or manipulation. The concept of IT infrastructure includes utilizing modern
technology to automate processes, improve efficiency, and enhance data accuracy. This
project aims to develop a Smart Attendance System using face recognition technology to
automate and optimize attendance management. The project will leverage AI-driven facial
recognition software to identify individuals and mark attendance automatically, ensuring a
seamless and secure process.

3. Objectives :
Develop a system that uses face recognition to automate attendance tracking.

Achieve high accuracy in identifying individuals using AI-driven facial recognition.

Ensure the system is secure and can handle real-time attendance monitoring.

Integrate the system with existing IT infrastructure for seamless data storage and retrieval.

Provide a user-friendly interface for administrators to manage attendance records.

4. Literature Review/Background Study :
Numerous studies have explored face recognition technologies, such as the use of OpenCV
and deep learning frameworks like TensorFlow and Keras. Existing attendance systems have
shown limitations in manual tracking and verification processes, highlighting the need for
automated solutions. Key references include:

"Face Recognition using Deep Learning" - Journal of Computer Science

OpenCV documentation and tutorials

5. Problem Statement
Traditional attendance systems are often time-consuming and subject to manipulation (e.g.,
buddy punching). The lack of automation leads to inaccuracies and administrative overhead.
This project addresses these challenges by providing a reliable and efficient face
recognition-based attendance system.

6. Proposed Solution/Methodology
The proposed solution involves developing a smart attendance system that uses facial
recognition to automatically identify individuals and mark their attendance. The
methodology includes:

Face Detection: Detect faces in live video streams using algorithms like Haar Cascades or
Dlib's face detector.

Face Recognition: Extract facial features using a pre-trained model (e.g., FaceNet) and
compare them with a stored database of registered users.

Attendance Marking: Automatically log attendance in a database if a recognized face is

detected.

User Interface: Develop an admin panel to view and manage attendance records.

Security Measures: Implement data encryption and secure storage practices to protect user
information.
7. Infrastructure Requirements
Hardware:

High-resolution camera for capturing facial images.

A server with GPU capabilities for training and running AI models.

Network infrastructure to connect cameras and the server.

Software:

Operating system (Windows, Linux).

Python and related libraries (OpenCV, Dlib, TensorFlow/Keras).

Database management system (MySQL, MongoDB).

Web server for the admin interface (Apache, Nginx).

Other Components:

Pre-trained facial recognition models (e.g., FaceNet).

8. Project Plan and Timeline

Week 1-2*: Requirement gathering and market research

- Week 3-4: Design and architecture planning

- Week 5-6: Infrastructure setup and backend development

- Week 7-8: Frontend development and module integration

- Week 9: Testing and user feedback

- Week 10: Final deployment and project review

9. Expected Outcomes

A functional smart attendance system that uses face recognition to automate attendance
tracking.

Achieve an accuracy rate of at least 95% for face recognition.

Real-time attendance monitoring with secure data storage.

User-friendly interface for attendance management.

Reduced time spent on manual attendance processes and improved data accuracy.

10. References

Schroff, F., Kalenichenko, D., & Philbin, J. (2015). FaceNet: A unified embedding for face
recognition and clustering. IEEE Conference on Computer Vision and Pattern Recognition.

Kazemi, V., & Sullivan, J. (2014). One millisecond face alignment with an ensemble of
regression trees. IEEE Conference on Computer Vision and Pattern Recognition.

OpenCV Library Documentation. (n.d.). Available at: https://docs.opencv.org

Submission Date :

Supervisor Sign :

Student Name : 1 . ANIKET CHOUDHARI(223580

2. GAURAV BORKAR(22358010046)

3.SUMIT ANDHARE(22358010006)

4 .LAXMAN CHOUDHARI(22358010052)

5.ATHARV BEDARKAR(22358010019)