ESP32-S3 Thermal Imaging Camera

This project implements a thermal imaging application using an ESP32-S3 development board (FeatherS3 from Unexpected Maker) and an MLX90640 thermal camera sensor. The application communicates with the sensor via I2C to capture thermal images and provides a console interface for controlling the camera and viewing the thermal data.

Hardware Requirements

• Unexpected Maker FeatherS3 development board (ESP32-S3)
• MLX90640 Thermal Camera Breakout (connected via STEMMA QT socket)
• USB cable for programming and serial communication

Wiring

Connect the MLX90640 thermal camera to the FeatherS3 board's STEMMA QT socket (LDO 1) using a 4-wire connection:

• VIN -> 3.3V
• GND -> GND
• SDA -> GPIO8 (I2C data line)
• SCL -> GPIO9 (I2C clock line)

Building and Flashing

Prerequisites

• ESP-IDF development environment set up according to Espressif's instructions
• ESP-IDF environment variables set (run . $HOME/esp/esp-idf/export.sh if needed)

Build and Flash

# Navigate to the project directory
cd ~/github/esp/thermal_camera

# Build the project
idf.py build

# Flash the project to the ESP32-S3
idf.py -p /dev/ttyACM0 flash

# Monitor the serial output
idf.py -p /dev/ttyACM0 monitor

Project Architecture

graph TD
    A[MLX90640 Thermal Camera] -- I2C --> B(ESP32-S3 FeatherS3);
    B -- USB Serial --> C[PC/Console Interface];
    B -- Wi-Fi --> D[Remote Monitoring Interface / App];
    B -- SPI/GPIO --> E[External Display];
    B -- SPI --> F[SD Card for Data Logging];
    B --> G{Image Processing & Filtering};
    G --> C;
    G --> D;
    G --> E;
    G --> F;

    subgraph "User Interaction"
        C
        D
    end

    subgraph "Data Storage & Display"
        E
        F
    end

    subgraph "Core Processing"
        B
        G
    end

    subgraph "Sensor"
        A
    end
end

Loading

Usage

The camera operates by wirelessly transmitting thermal images. Currently, it streams video continuously. The plan is to modify this to capture and transmit a still thermal image approximately every 2.5 seconds.

The primary interaction is wireless, and the previous serial console command-line interface is no longer the main method of operation.

Example Output

When running the capture command, you'll see output similar to this:

Thermal Data:
Min temperature: 22.50°C
Max temperature: 28.75°C
Sample temperatures (°C):
24.3 23.8 23.5 24.1 24.6 25.0 25.3 25.1 
23.9 23.5 23.8 24.3 25.1 25.7 26.2 25.8 
24.1 24.0 24.5 25.2 26.3 27.1 27.5 26.9 
24.8 24.9 25.5 26.4 27.5 28.2 28.5 27.8 
25.2 25.4 26.1 27.0 28.0 28.7 28.6 27.9 
24.9 25.1 25.8 26.5 27.3 27.8 27.5 26.8 

In continuous mode, the application displays a simple ASCII representation of the thermal image, updating in real-time.

Customization

You can modify the following parameters in the code to match your specific setup:

• I2C GPIO pins: If your MLX90640 is connected to different pins, update the I2C_MASTER_SDA_IO and I2C_MASTER_SCL_IO values in thermal_camera_main.c
• Default settings: You can change the default resolution and refresh rate in thermal_camera_main.c

Troubleshooting

• If the sensor is not detected, check your wiring connections and ensure the I2C pins are correctly configured
• Make sure the MLX90640 is properly powered (3.3V)
• Check that the I2C address matches (default is 0x33)
• Try reducing the I2C frequency if communication is unstable

Future Enhancements

• Add support for external display to show thermal images
• Implement data logging to SD card
• Add more advanced image processing and filtering
• Implement Wi-Fi connectivity for remote monitoring

License

This project is open source and available under the MIT License.# esp32-s3-thermal-camera-