Campus Crowd Density Awareness System (V1.0)

This project implements a real-time crowd density monitoring system using a single laptop. It consists of three components:

1. Edge Service: Captures video from a webcam/IP camera, detects people using YOLOv8, and sends data to the backend.
2. Backend Service: A FastAPI server that receives crowd data and serves it to the frontend.
3. Frontend Service: A web dashboard that displays real-time crowd density, count, and trends.

---

🛠️ Prerequisites

• Python 3.9+ installed.
• Git installed.
• Webcam (for testing) or Wi-Fi IP Camera (for demo).

---

🚀 Installation

1. Clone the Repository

   git clone https://github.com/falcon370/design_thinking.git
   cd design_thinking

2. Create a Virtual Environment

   python -m venv venv

3. Activate the Virtual Environment

   • Windows (PowerShell):

     .\venv\Scripts\activate

   • Mac/Linux:

     source venv/bin/activate

4. Install Dependencies

   pip install -r requirements.txt

---

▶️ How to Run

Option 1: Automatic (Recommended for Windows)

We have provided a PowerShell script to start all services in separate windows.

1. Open a PowerShell terminal in the project root.
2. Run the script:
   • Localhost (Default):

     .\start_system.ps1

   • Specific IP (e.g., for mobile testing): If you want to access the dashboard from your phone or another computer, use your laptop's Wi-Fi IP address (e.g., 192.168.68.104).

     .\start_system.ps1 -HostIP "192.168.68.104"

   • Optional: Run with debug logging enabled:

     .\start_system.ps1 -LogLevel DEBUG

3. The dashboard will be available at:
   • Localhost: http://localhost:8081
   • IP Mode: http://<YOUR_IP>:8081 (e.g., http://192.168.68.104:8081)
4. A window showing the camera feed with detections will appear.

Option 2: Manual Startup

If you prefer to run services manually, open three separate terminals and follow these steps:

Terminal 1: Backend Service

.\venv\Scripts\activate
$env:LOG_LEVEL='INFO'; uvicorn src.backend_service.main:app --host 0.0.0.0 --port 8000 --reload

Terminal 2: Frontend Service

python -m http.server 8081 --directory src/frontend

Open http://localhost:8081 in your browser.

Terminal 3: Edge Service

.\venv\Scripts\activate
# Usage: python src/edge_service/main.py <location_id> <camera_source>
python src/edge_service/main.py canteen 0

---

⚙️ Configuration

Multi-Camera Support

You can run multiple instances of the Edge Service for different locations.

# Terminal A (Canteen - Webcam)
python src/edge_service/main.py canteen 0

# Terminal B (Library - IP Camera)
python src/edge_service/main.py library "rtsp://192.168.1.105:554/stream1"

Logging & Debugging

The system supports three log levels, configurable via the LOG_LEVEL environment variable or the -LogLevel script parameter:

• INFO (Default): Standard startup/shutdown and transmission logs.
• DEBUG: Detailed logs including every detection count per frame.
• CRITICAL: Only errors and critical failures.

🌍 Running on Local Network (Access from Mobile)

To access the dashboard from other devices (e.g., a phone) on the same Wi-Fi:

1. Find your Laptop's IP Address:

   • Open PowerShell and run ipconfig.
   • Look for "IPv4 Address" (e.g., 192.168.1.10).

2. Update Frontend Config:

   • Open src/frontend/index.html.
   • Change line 55:

     // const API_URL = "http://localhost:8000/current-status";
     const API_URL = "http://192.168.1.10:8000/current-status"; // Replace with YOUR IP

3. Restart Frontend:

   • If running manually, restart the frontend terminal.
   • If using start_system.ps1, close and run it again.

4. Access from Mobile:

   • Open browser on phone: http://192.168.1.10:8081

---

📂 Project Structure

/src
  /backend_service    # FastAPI Server (main.py, test_backend.py)
  /edge_service       # Computer Vision Logic (main.py)
  /frontend           # Web Dashboard (index.html)
requirements.txt      # Python dependencies
start_system.ps1      # Startup script
README.md             # This file

🛑 Troubleshooting

• "Module not found": Ensure you activated the virtual environment (.\venv\Scripts\activate) before running python commands.
• Camera not opening: Check if another app (Zoom, Teams) is using the camera. Verify CAMERA_SOURCE in src/edge_service/main.py.
• Backend connection error: Ensure the Backend Service is running on port 8000 before starting the Edge Service.