A real-time library occupancy tracking system using AI-powered object detection to monitor cubicle availability.
LibraTrack uses TensorFlow.js and the COCO-SSD model to detect people in library cubicles, providing real-time occupancy status. The system processes camera feeds every 5 seconds to determine which seats are available or occupied.
- Real-time Object Detection - Uses TensorFlow.js COCO-SSD model for person detection
- Configurable Detection Zones - Define custom crop zones for each cubicle
- Visual Status Dashboard - Color-coded indicators (green = available, red = occupied)
- Live Camera Feed - Monitor library spaces in real-time
- Session Management - Track occupancy patterns over time
- React - Frontend framework
- TensorFlow.js - Machine learning library
- COCO-SSD - Pre-trained object detection model
- React Webcam - Camera integration
- React Image Crop - Zone configuration
- Node.js (v14 or higher)
- npm or yarn
# Install dependencies
npm install
# Start development server
npm startThe app will open at http://localhost:3000
npm run build- Configure Zones - Define detection areas for each cubicle using the Configuration tab
- Camera Feed - The system captures frames from the camera feed
- Object Detection - TensorFlow.js analyzes each zone for person detection
- Status Update - Occupancy status updates every 5 seconds
- Visual Display - Dashboard shows real-time availability
src/
├── Components/ # React components (Navbar, Seating, etc.)
├── Contexts/ # React context for state management
├── assets/ # Images and static files
├── imageprocessor.js # Image processing utilities
├── utilities.js # Helper functions
└── App.js # Main application component
- Navigate to the Configure Detection Zones tab
- Upload or select your camera feed
- Draw crop zones around each cubicle
- Save the configuration
- Switch to the Dashboard tab to view real-time occupancy
This project is licensed under the MIT License.