techn0man1ac / AppleDetectAISystem

Apple Detect AI System is a simple AI-powered system that detects apples using a camera and controls an Arduino board based on the detection result. This project uses LM Studio to run the Gemma 3 4b multimodal AI model locally.

Apple Detect AI System

Apple Detect AI System is a simple AI-powered system that detects apples using a camera and controls an Arduino board based on the detection result. This project uses LM Studio to run the Gemma 3 4b multimodal AI model locally.

Video demonstration

https://youtube.com/shorts/DjQJwr3t2xY

How It Works

1. The camera captures an image.
2. The image is sent to the AI model Gemma 3 4b running on LM Studio.
3. The AI analyzes the image and determines if an apple is present.
4. If an apple is detected:
   • The system sends 1 to the Arduino board.
   • The Arduino turns on an LED (or another connected device).
5. If no apple is detected:
   • The system sends 0 to the Arduino board.
   • The Arduino turns off the LED.

Requirements

Hardware

• Computer (to run LM Studio and process images)
• Camera (compatible with pygame.camera)
• Arduino Board (e.g., Arduino Uno)
• LED or other output device
• USB cable (to connect the Arduino)

Software

• Python 3
• LM Studio (running Gemma 3 4b)
• Arduino IDE
• Python Libraries:
  • openai
  • pygame
  • pyserial

Installation & Setup

1. Set Up LM Studio

• Download and install LM Studio.
• Load the Gemma 3 4b model.
• Start the LM Studio server (http://localhost:1234/v1).

2. Install Python Dependencies

pip install openai pygame pyserial

3. Upload Arduino Code

• Open SaAIDS.ino in the Arduino IDE.
• Select the correct board and port.
• Upload the code to the Arduino.

4. Run the Python Script

python CVTestAI.py

Expected Output

• If an apple is detected: LED turns on.
• If no apple is detected: LED remains off.
• The console prints: Yes or No.

License

This project is licensed under the MIT License.