ESP32-CAM Stream Vision

A flexible real-time camera streaming system with OpenCV integration for computer vision projects. This repository contains both the ESP32-CAM firmware and Python client code for processing the camera stream.

Features

• Dual Network Operation: Connect to a WiFi hotspot while maintaining fallback AP mode
• Camera Stream Endpoint: Access raw camera feed for computer vision projects
• Web Dashboard: Monitor sensors and camera feed through an intuitive interface
• OpenCV Integration: Python client for capturing and processing the video stream
• REST API: Access sensor data and system information programmatically
• Fallback Mechanisms: Automatic recovery from network interruptions

Hardware Requirements

• ESP32-CAM module (AI-Thinker ESP32-CAM recommended)
• Optional: Additional sensors (DHT22, water level sensor, etc.)
• A computer or mobile device to view the stream and run OpenCV code

Repository Structure

ESP32-CAM-Stream-Vision/
├── firmware/
│   └── ESP32_CAM_Dual_Network/
│       └── ESP32_CAM_Dual_Network.ino
├── client/
│   ├── esp32_opencv.py
│   └── advanced_processing.py
├── docs/
│   ├── images/
│   └── api_reference.md
└── README.md

Getting Started

1. Upload the Firmware

1. Open firmware/ESP32_CAM_Dual_Network/ESP32_CAM_Dual_Network.ino in the Arduino IDE
2. Update the WiFi credentials:

// WiFi client mode settings (your hotspot credentials)
const char* wifi_ssid = "YOUR_HOTSPOT_SSID";
const char* wifi_password = "YOUR_HOTSPOT_PASSWORD";

3. Select the correct board (ESP32 AI-Thinker CAM)
4. Upload the code to your ESP32-CAM

2. Access the Web Interface

The ESP32-CAM creates its own access point by default, but will also try to connect to your specified network:

• Access Point Mode: Connect to the WiFi network ESP32-CAM-AP with password 12345678
• Client Mode: The ESP32-CAM will also connect to your specified WiFi network

Navigate to one of these addresses in your web browser:

• AP Mode: http://192.168.4.1
• Client Mode: Check the Serial Monitor output for the assigned IP address

3. Run the Python Client

The Python client requires OpenCV and can capture and process the camera stream:

# Install dependencies
pip install opencv-python numpy

# Run the client
python client/esp32_opencv.py --ip 192.168.X.X  # Replace with your ESP32-CAM's IP

Available Endpoints

The ESP32-CAM provides several HTTP endpoints:

Endpoint	Method	Description
/	GET	Main dashboard with JavaScript-based updates
/auto	GET	Simple dashboard with meta tag auto-refresh (3s)
/test	GET	Minimal test page with auto-refresh
/stream	GET	MJPEG camera stream (for browsers and OpenCV)
/sensor-data	GET	JSON API with current sensor readings
/update-sensor	POST	Endpoint to update sensor data
/network-status	GET	Network information page

OpenCV Integration

The Python client demonstrates basic image processing with OpenCV:

• Edge detection
• Frame manipulation
• Basic computer vision

Modify process_frame() in esp32_opencv.py to implement your own computer vision algorithms.

Example Projects

This platform can be used for various applications:

• Home security and surveillance
• Plant growth monitoring
• Water level detection
• Motion detection and tracking
• Object recognition (with additional ML models)

Troubleshooting

• Cannot connect to ESP32-CAM: Try the AP mode first (192.168.4.1)
• Stream is slow or unstable: Reduce the frame size in the ESP32-CAM code
• Camera not detected: Check connections and power supply
• Python client crashes: Ensure OpenCV is installed correctly

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• ESP32-CAM examples by Espressif
• OpenCV library contributors