Abstract :

This Paper Introduces An Efficient And Costeffective Web-based Face Recognition Attendance System Designed To Automate Attendance Management While Simultaneously Analyzing Student Behavior In Real Time. The Proposed System Incorporates Six Monitoring Indicators, Including Present, Absent, Late Arrival, Early Departure, Proxy Attendance, And Carelessness Activities Such As Mobile Phone Usage And Sleeping During Lectures. To Improve Detection Speed And Recognition Accuracy, The Framework Integrates A Hybrid Architecture Combining The Classical Viola–Jones Cascade Algorithm For Rapid Frontal Face Detection With Advanced Deep Learning Models, Namely YOLOv8n-face And ArcFaceResNet100. The Developed System Operates Effectively On Standard Laptop Hardware Using Only Two 1080p Cameras, Making It Suitable For Practical Classroom Deployment With Minimal Infrastructure Cost. Experimental Evaluation Was Conducted On 115 Students Across 1,240 Classroom Sessions. The Results Demonstrated A Recognition Accuracy Of 99.41% With Real-time Processing Capability Of 42 Frames Per Second. In Addition, The Framework Successfully Achieved Complete Proxy Attendance Identification And 97.2% Accuracy In Detecting Careless Student Activities. The System Further Enhances Administrative Efficiency Through Automatic Generation Of Attendance Reports In Excel And Google Sheets Formats, Along With Instant WhatsApp Notifications For Monitoring And Communication Purposes. The Proposed Solution Provides A Reliable, Scalable, And Intelligent Attendance Management Platform Suitable For Modern Educational Environments. Keywords: Face Recognition Attendance System, Viola–Jones, YOLOv8, ArcFace, Proxy Detection, Carelessness Detection.

Published:

12-5-2026

Issue:

Vol. 26 No. 5 (2026)

Page Nos:

1087-1092

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite