Ultimate Computer Vision TicTacToe (U-CV-TTT)

Overview

Ultimate Computer Vision TicTacToe is a Python project that simulates a TicTacToe player whose "eyes" are the camera of a mobile device. The application uses Python and OpenCV to recognize and interact with the TicTacToe board.

Key Features

• Play against an AI with three difficulty modes: Easy, Medium, and Nightmare.
• Real-time board recognition via a mobile device camera.
• Train and test your own symbol classifier for custom configurations.

Requirements

• Python (latest version recommended)
• DroidCam installed on your mobile device

Installation

Follow these steps to set up the project:

1. Clone the Repository

   git clone https://github.com/macorisd/U-CV-TTT.git

2. Install Python

   • Go to python.org/downloads.
   • Download and install the latest version of Python.

3. Install DroidCam on a Mobile Device

   • For Android: Download from Google Play Store.

4. Install Required Python Packages

   • Open the project folder, named U-CV-TTT.

   • Right-click in the folder and select Open in Terminal.

   • Install dependencies using one of the following methods:

     Option 1: Run the batch file install_packages.bat located in the Setup folder.

     Option 2: Install packages manually by running the following commands:

     pip install opencv-python
     pip install shapely
     pip install matplotlib

5. You're ready to play!

   • Open the file U-CV-TTT.pyw for direct execution, or run U-CV-TTT.py to view the Python code.

How to Play

1. Connect both your mobile device and computer to the same Wi-Fi network (or tether your computer to your mobile data).
2. Launch DroidCam on your mobile device.
3. Run the U-CV-TTT.pyw file.
4. Click on the IP/Port button in the application window.
5. Enter the IP and port provided by DroidCam into the corresponding fields.
6. Select a difficulty mode: Easy, Medium, or Nightmare.
7. Capture board images by pressing the spacebar. Exit the application by pressing q.

Tips for Optimal Performance

• Use plain white paper and a black marker for better recognition.
• Align the TicTacToe board approximately within the provided grid, but avoid placing it too close to the camera.
• You may choose to draw or not draw the bot's moves on the paper; both approaches work.

Advanced Features

1. Train Your Own Symbol Classifier

   • Use the train_classifier.py script to train a custom symbol classifier.

2. Test the Symbol Classifier

   • Use the test_classifier.py script to evaluate your classifier's performance.

Author

Made with ❤️ by Macorís Decena Giménez.
GitHub Profile

Reporting Issues

If you encounter any bugs or issues, please feel free to contact me at macorisd@gmail.com.