AI Tic-Tac-Toe with Computer Vision and CLI

A Python project offering multiple interfaces to play Tic-Tac-Toe against AI opponents. Features terminal gameplay with two AI strategies (Minimax and MCTS) and computer vision integration for real-world board analysis.

Features

Core Game Modes

• 🎮 Terminal Gameplay: Play classic Tic-Tac-Toe in your terminal against AI
• 🤖 Dual AI Engines: Choose between:
  • Minimax with alpha-beta pruning (perfect play)
  • Monte Carlo Tree Search (MCTS) opponent
• 📸 Computer Vision Integration: Process physical/drawn boards from images
  • Automatic board state detection
  • Optimal move visualization on input images

Technical Capabilities

• 🖼️ Image Processing Pipeline:
  • Grid detection with OpenCV
  • Cell segmentation and symbol recognition
• 🧠 AI Decision Systems:
  • Minimax (integrated with CV pipeline)
  • MCTS (terminal-only implementation)
• ✍️ Visualization Tools:
  • Move annotation on processed images
  • Terminal-based board display

Requirements

• Python 3.8+
• OpenCV 4.5+
• numpy 1.20+

Installation

git clone https://github.com/mzums/tic-tac-toe.git
cd tic-tac-toe
pip install -r requirements.txt

Usage

Terminal Game Mode

python <minimax_or_mcts>/main.py

Computer Vision Mode

python main.py path/to/image.jpg

Options:

• --show-steps to display processing stages

Test images:

avaible in folder cv/test_images

usage:

python main.py cv/test_images/image1.jpg --show_steps

Image Requirements

For best CV performance:

• High contrast between grid and background
• Uniform lighting
• Square board proportions
• Supported formats: JPG, JPEG, PNG, BMP

License

MIT License. See LICENSE for details.

Contributing

1. Fork repository
2. Create feature branch (git checkout -b feature/improvement)
3. Commit changes
4. Push to branch
5. Open Pull Request