ryukaizen/facelogger

A face recognition check-in module based on AI-Thinker's ESP32-CAM using C++ and Python.

Facelogger

A face recognition check-in module based on AI-Thinker's ESP32-CAM

This project combines the functionalities of ESP32-CAM with Python's face-recognition library for detecting and identifying faces in real time.

📚 Table of Contents

• 🛠 ESP32 Board Setup
• 🔌 Hardware Setup
• 💻 ESP32-CAM Programming
• 📥 Installation
• 🚀 Usage
• 🤝 Contributing
• 📄 License

🛠 ESP32 Board Setup

Before proceeding with the hardware connection, ensure your Arduino IDE is properly configured for ESP32 development:

1. Configure Additional Board Manager URL:

   • Navigate to "File" > "Preferences" (on macOS, "Arduino" > "Preferences").
   • Locate the "Additional Boards Manager URLs" field.
   • Click the icon next to the field to open the input window.
   • Add the following URL on a new line:

     https://dl.espressif.com/dl/package_esp32_index.json
     

   • Click "OK" to save the changes and close the Preferences window.

2. Install ESP32 Board Support:

   • Open the Boards Manager by selecting "Tools" > "Board" > "Boards Manager".
   • In the search bar, type "esp32".
   • Locate "esp32 by Espressif Systems" and click "Install".
   • After installation, close the Boards Manager.

3. Select ESP32 Board:

   • Go to "Tools" > "Board" and select your specific ESP32 board model from the list.

After completing these steps, your Arduino IDE will be ready for ESP32 development.

🔌 Hardware Setup

ESP32-CAM and FTDI Programmer Connection

Before programming, connect your ESP32-CAM to an FTDI programmer as follows:

ESP32-CAM	FTDI Programmer
GND	GND
5V	VCC
U0R	TX
U0T	RX
GPIO0	GND

⚠️ Important: Short GPIO0 to GND to enter programming mode. Remove this connection after programming.

OLED Display Connection

We have used OLED SSD1306 128x64 to display various statuses.

After uploading the code to ESP32-CAM:

1. Remove the GPIO0 to GND connection.
2. Connect OLED GND to ESP32-CAM GND.
3. Connect OLED VCC to ESP32-CAM 3V.
4. Press the reset button on ESP32-CAM.

The OLED should display "ESP32 Started!" 🎉

💻 ESP32-CAM Programming

1. Navigate to the aithinker/src/ folder.
2. Open esp32cam.ino in Arduino IDE.
3. Add your Wi-Fi credentials:

static const char* WIFI_SSID = "your_wifi_name";
static const char* WIFI_PASS = "your_wifi_password";

4. Upload the code to ESP32-CAM (ensure GPIO0 is connected to GND).
5. After uploading, disconnect GPIO0 from GND and reset the board.
6. Open the Serial Monitor to get the generated link.

📥 Installation

1. Clone this repository:

git clone https://github.com/ryukaizen/facelogger.git
cd facelogger

2. Set up the ESP32-CAM as described in the Hardware Setup section.
3. Install Python dependencies:

pip install -r requirements.txt

Before running the project, you need to set up some key variables in the main.py file:

# In main.py

# Replace this with your own ESP32-CAM generated link (check aithinker folder)
ESP32_CAM_URL = 'http://192.168.52.160/640x480.jpg'

# Replace this with your own Google Form URL
GOOGLE_FORM_URL = "https://docs.google.com/forms/d/e/AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz/formResponse"

# Replace this with your own MongoDB URL
MONGODB_URL = 'mongodb://localhost:27017/'

# Inspect element on the Google Form and find the entry name ID and entry time ID
ENTRY_NAME_ID = 'entry.123456789' 
ENTRY_TIME_ID = 'entry.987654321'

Ensure you replace these placeholder values with your actual URLs and IDs:

1. ESP32_CAM_URL: The URL generated by your ESP32-CAM. You can find this in the Serial Monitor after uploading the code to your ESP32-CAM.

2. GOOGLE_FORM_URL: The URL of your Google Form where names and timestamps will be logged.

3. MONGODB_URL: The URL of your MongoDB instance.

4. ENTRY_NAME_ID and ENTRY_TIME_ID: These are specific to your Google Form. To find these:

   • Open your Google Form in a web browser
   • Right-click and select "Inspect" or "Inspect Element"
   • Look for input fields with names like "entry.123456789"
   • The number after "entry." is your ID

⚠️ Note: Keep your Google Form URL and entry IDs private to prevent unauthorized access to your logging data.

🚀 Usage

1. Program the ESP32-CAM and note the generated URL.
2. Add face images into the training_data folder. Make sure those are in .jpeg format and are of good quality.
3. Run the Python script:

python3 main.py

The system will now detect and recognize faces using the ESP32-CAM feed.

⚠️ Note: For the system to work properly, both the ESP32-CAM and the computer running the Python script must be on the same local network.

🤝 Contributing

As always, contributions are welcome! Feel free to submit a PR. You can reach out to me on Telegram.

📄 License

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

---

Made with ❤️ by ryukaizen