Face Mask Detection System
Real-time mask compliance monitoring system with 95%+ accuracy for 3-class detection.
Project Overview
A factory-floor safety monitoring system detecting mask-worn, mask-off, and incorrect mask usage in real time. Works on live webcam with compliance rate tracking and color-coded alerts. MobileNetV2 achieves fast inference on CPU.
⚠ The Problem
Manufacturing facilities and public spaces needed automated mask compliance monitoring during health regulations, without expensive GPU hardware.
✓ The Solution
MobileNetV2 transfer learning with 2-phase fine-tuning. Data augmentation for robustness across lighting and angles. Streamlit dashboard with real-time webcam detection and compliance reporting.
Results & Outcomes
95%+ accuracy across all 3 classes. Real-time webcam detection with person-level compliance tracking.