MTGTextRemover

Removes specific text (e.g., watermark phrases) from Magic: The Gathering card images. Uses Tesseract OCR to detect targeted words or phrases and OpenCV inpainting to remove them.

Features

• Detects user-defined text via Tesseract.
• Configurable bounding box padding, dilation, and inpainting parameters.
• Supports multi-word phrases.
• Preserves or changes the image extension as needed.
• Offers a GUI (using CustomTkinter) for non-technical users.

Requirements

• Python 3.8+
• Tesseract installed system-wide or specify a custom path.
• Python packages:
  • opencv-python
  • pytesseract
  • PyYAML
  • customtkinter (for the GUI)

Installation

1. Install Tesseract (via your OS package manager or official sources).
2. Clone or download this repository.
3. Install Python dependencies:

   pip install opencv-python pytesseract PyYAML customtkinter

Usage (GUI)

1. Run the gui.py script:

   python gui.py

2. Click Open Config to load a YAML file (optional).
3. Set input/output directories, text phrases, and inpainting settings.
4. Click Run to process images.

Usage (CLI)

If you have a CLI version (e.g., main.py):

1. Edit or use default config (config.yaml).
2. Run:

   python main.py --config-file config.yaml

3. Images are read from the specified input folder and saved to the output folder.

Configuration

A YAML config can include phrases, bounding box parameters, inpainting radius and method, Tesseract path, and more. See config.yaml as an example.

---

Disclaimer
This software is for personal use and made with ChatGPT. Use at your own risk.