SMART HEALTH MONITORING SYSTEM

An real time digital wellness system that monitors eye blinks, posture and screen time behaviour using computer vision and IoT principles to promote healthier digital habits.

Problem Statement

Prolonged screen exposure leads to:

• Eye strain
• Poor posture
• Fatigue
• Reduced productivity Most solutions are either expensive wearables or simple reminder apps. This project provides an integrated, low-cost, real-time monitoring system using a webcam and AI-based detection.

System Architecture

The system follows a modular, real-time architecture:

• Data Acquisition Layer
  • Webcam video capture (30 FPS)
  • Frame preprocessing
• Processing Layer
  • Blink Detection (MediaPipe Face Mesh + EAR calculation)
  • Posture Detection (MediaPipe Pose)
  • Alert Management System
• Storage & Reporting Layer
  • Local CSV logging
  • Firebase Realtime Database sync
  • Health Report Generator

Tech Stack

• Language
  • Python 3.10+
• Computer Vision & AI
  • OpenCV
  • MediaPipe
  • TensorFlow Lite
• Data Handling
  • Pandas
  • CSV logging
  • Firebase Realtime Database
• UI & Notifications
  • Tkinter / CustomTkinter
  • Plyer / Win10Toast

Key Features

• Real-time blink detection using Eye Aspect Ratio (EAR)
• Posture angle monitoring with slouch detection
• Instant alerts for drowsiness and bad posture
• Health dashboard with screen-time insights
• Cloud synchronization (Firebase)
• Automated report generation
• Privacy-focused local data processing

Core Detection Logic

• Blink Detection
  • Uses MediaPipe Face Mesh
  • Computes Eye Aspect Ratio (EAR)
  • Triggers alert if eyes closed > 2 seconds
• Posture Detection
  • Tracks neck and torso landmarks
  • Calculates neck angle
  • Triggers alert if angle > 25° for > 2 seconds

File Structure

├── main.py
├── blink_detector.py
├── posture_detector.py
├── data_logger.py
├── firebase_logger.py
├── health_report.py
├── ui_notifications.py
└── requirements.txt

Installation

git clone <repo-link>
cd smart-health-monitor
pip install -r requirements.txt
python main.py

Future Improvements

• AI-based predictive fatigue scoring
• Mobile companion app
• Telemedicine integration
• Enhanced posture calibration
• Multi-sensor integration (ECG, SpO₂)