Scribd
Upload
9 views13 pages

Edunet

The document presents a real-time attendance monitoring system using face recognition, addressing the inefficiencies of manual attendance methods. It utilizes a webcam, OpenCV, and Flask to automate attendance logging with high accuracy and a user-friendly interface. The proposed system aims to enhance reliability and reduce manual workload while outlining future improvements such as advanced algorithms and mobile integration.

Uploaded by

sreeshanthkumarnookala
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
9 views13 pages

Edunet

The document presents a real-time attendance monitoring system using face recognition, addressing the inefficiencies of manual attendance methods. It utilizes a webcam, OpenCV, and Flask to automate attendance logging with high accuracy and a user-friendly interface. The proposed system aims to enhance reliability and reduce manual workload while outlining future improvements such as advanced algorithms and mobile integration.

Uploaded by

sreeshanthkumarnookala
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

Real-Time

Attendance
Monitoring with
Face Recognition
and Flask

PRESENTED BY

STUDENT NAME: N.SREESHANTH


KUMAR

COLLEGE NAME:MGIT

DEPARTMENT:ECE

EMAIL ID:[email protected]

AICTE STUDENT ID: AINSI_113877


OUTLINE
• Problem Statement
• Proposed System/Solution
• System Development Approach
• Algorithm & Deployment
• Result
• Conclusion
• Future Scope
• References
PROBLEM STATEMENT
Manual attendance systems can be time-consuming, often made wrong
by human errors, and can lead to proxy attendance (buddy punching),
leading to inaccuracies in data and productivity in educational and
organizational environments. Biometric methodologies (finger print
scanning, etc.) have problems with hygiene, as was the case with the pre
COVID systems, people needed to touch the same spot to scan, and
during COVID that was more than problematic to return to the earlier
biometrics.

We need a non-intrusive, automated, reliable attendance system that will


minimize the manual workload associated with attendance, limit
dishonest attendance, and allow the logging of data in real-time.
PROPOSED SOLUTION
To address the challenges of traditional
attendance systems, we propose a real-time
To address the challenges of traditional attendance systems, we propose a real-time face recognition-
basedface recognition-based
attendance system usingattendance
a webcam, system
OpenCV, and a Flask web interface. This system detects and
recognizes student faces
using a webcam, fromand
OpenCV, liveavideo,
Flask records
web attendance instantly with timestamp and ID, and
updates it in a CSV file. It eliminates manual errors, ensures faster data recording, and maintains
interface. This system detects and recognizes
accurate logs — all accessible through a user-friendly web dashboard.
student faces from live video, records
attendance instantly with timestamp and ID,
and updates it in a CSV file. It eliminates
manual errors, ensures faster data recording,
and maintains accurate logs — all accessible
through a user-friendly web dashboard.
SYSTEM APPROACH
🖥️Hardware
Component Specification

Webcam Built-in or external

RAM Minimum 4 GB

Processor Intel i3 or above

💻 Software
Component Specification

OS Windows / Linux / macOS

Python Version 3.x

Web Browser For accessing Flask interface

Code Editor VS Code / PyCharm


✅ 2. Required Python Libraries

Library Use / Function

OpenCV Face detection, image processing

Flask Web app framework (interface backend)

NumPy Numerical operations and array handling

Pandas Reading/writing CSV files (attendance logs)

Joblib Save/load trained ML models

Scikit-learn K-Nearest Neighbors (KNN) classifier implementation

Datetime Timestamp generation for attendance

OS Directory and file management


ALGORITHM & DEPLOYMENT
🔍 Algorithm Selection
 The system uses K-Nearest Neighbors (KNN) classifier.

 KNN was selected for its simplicity, speed, and high accuracy in pattern recognition tasks such as face
recognition.

 It works well with low to moderate-sized datasets, which suits our attendance system.

📥 Data Input
 The input to the algorithm includes:

 Facial images captured via webcam.

 Preprocessed images resized to uniform dimensions.

 Flattened pixel arrays as input vectors.


🏋️Training Process
 Collected facial images for each registered user.
 Images are resized and converted into feature vectors.
 Model is trained using KNN to map feature vectors to user labels

(Name_ID).
 Joblib is used to save the trained model for reuse.

📊 Prediction Process
 The trained KNN model compares new face input with stored faces.
 It finds the nearest match and returns the predicted user label.
 Upon identification, the user’s attendance is marked with a

timestamp.
RESULT
🎯 Accuracy and Effectiveness
 The trained KNN face recognition model achieved over 90%
accuracy in identifying faces under
properlighting and camera angles.
 Real-time performance demonstrated:
• Fast detection (<1 second) per frame.
• Accurate identification even with minor variations in
expressions or angles.
• False positive rate: Low, due to clear training images and
preprocessed input.
📈 Visual Feedback
 Attendance logs are successfully recorded with:
• Name , Roll Number & Timestamp
• The results are displayed live on the Flask web interface after
each session.
CONCLUSION
This face recognition attendance system offers a fast, reliable, and contactless way to mark

attendance using real-time video. It reduces manual work and prevents proxy entries.
Highlights:
• Accurate face detection and recognition
• Easy-to-use web interface
• Real-time attendance logging

Challenges:
• Poor lighting and camera angle issues
• Limited training data
FUTURE SCOPE
• Multi-City Deployment: Expand the system to support institutions across different regions.

• Advanced Algorithms: Integrate deep learning models (e.g., CNNs) for improved face recognition accuracy.

• Edge Computing: Use edge devices (e.g., Raspberry Pi) to process data locally and reduce latency.

• Additional Data Sources: Incorporate student schedules, location data, or biometric sensors for smarter
verification.

• System Optimization: Enhance performance through better training techniques and real-time model
updates.

• Mobile & Cloud Integration: Support mobile-based check-ins and cloud dashboards for centralized
monitoring.
REFERENCES
• OpenCV Documentation – Face Detection using Haar Cascades
https://docs.opencv.org

• Flask Documentation – Web Framework for Python


https://flask.palletsprojects.com

• scikit-learn – Machine Learning in Python


https://scikit-learn.org

• Pandas Documentation – Data Analysis Library


https://pandas.pydata.org
Thank you

You might also like