Mistral OCR GUI Processor

A user-friendly desktop application, built with Python and Tkinter, to perform Optical Character Recognition (OCR) on images using the powerful Mistral AI API. The tool provides a simple interface to process individual images, entire folders, or nested subfolders, and can combine the extracted text into organized markdown files.

Features

• Intuitive Graphical User Interface: A clean, tabbed interface for different processing modes.
• Multiple Processing Modes:
  • Individual Images: Process one or more specific image files.
  • Folder Processing: Process all supported images within a single folder.
  • Subfolder Processing: Process all supported images within the immediate subdirectories of a parent folder.
• Drag-and-Drop Support: Easily add files or folders by dragging them onto the application window.
• Markdown Combination: Automatically combine OCR results from multiple images into a single, well-structured markdown file.
• Flexible Sorting Options: When combining files, sort them by natural filename order or by modification date.
• Real-time Progress Visualization: A grid of squares visually represents the status of each image (Processing, Completed, Error).
• Cancellable Operations: Stop a long-running processing job at any time.
• Concurrent Processing: Uses a thread pool to process multiple images in parallel, speeding up large jobs.

Requirements

• Python 3.8 or higher.
• A Mistral AI account and an API key. You can get one from the Mistral AI Platform.

Installation

The recommended way to install the application is using pip:

pip install mistral-ocr-gui

This command will download the application and automatically install all required dependencies.

Installation from Source

If you want to contribute to the project or install the latest development version, you can install it from the source code:

1. Clone the repository:

   git clone https://github.com/Danielnara24/Mistral-OCR.git
   cd Mistral-OCR
   

2. Install in editable mode: This command will install the application and its dependencies. The -e flag allows you to make changes to the source code and have them take effect immediately.

   pip install -e .
   

Configuration: Setting the API Key

The application requires your Mistral API key to be set as an environment variable named MISTRAL_API_KEY.

Windows

Open Command Prompt or PowerShell and run the following command to set the variable permanently. You will need to restart your terminal for the change to take effect.

setx MISTRAL_API_KEY "your_api_key_here"

macOS / Linux

Add the following line to your shell's configuration file (e.g., ~/.bashrc, ~/.zshrc, or ~/.profile):

export MISTRAL_API_KEY="your_api_key_here"

Then, either restart your terminal or run source ~/.bashrc (or the relevant file) to apply the changes.

Verify the Setup

You can verify that the environment variable is set correctly by running:

• (macOS/Linux): echo $MISTRAL_API_KEY
• (Windows CMD): echo %MISTRAL_API_KEY%
• (Windows PowerShell): echo $env:MISTRAL_API_KEY

Usage

Once installed, the application can be launched directly from your terminal. Simply run the following command:

mistral-ocr

The graphical user interface will appear, and you can start processing your images.

How It Works

1. Select a Tab:

   • Individual Images: Use this for processing a specific set of image files from different locations. Each image will generate a corresponding _OCR.md file in its original directory.
   • Folder: Use this to process all images inside a single folder. You have the option to combine all the results into one markdown file named Combined_OCR_[FolderName].md.
   • Subfolders: Use this to process images located in the immediate subfolders of a parent directory. You can combine results for each subfolder individually and even create a final compilation of all subfolder results.

2. Add Files/Folders:

   • Click the "Select..." button to open a file/folder dialog.
   • Or, drag and drop your files/folders directly onto the application window.

3. Configure Settings (if applicable):

   • For Folder and Subfolder modes, check the box to enable combining markdown files.
   • Choose a sorting method for the combined document.

4. Process:

   • Click the "Process" button to start the OCR job.
   • The progress grid will update in real-time, showing the status of each file.
   • A timer will show the elapsed time.

5. Output:

   • The generated markdown files (.md) will be saved in the same directory as the source images or in the relevant parent/subfolder directory for combined files.

License

This project is licensed under the MIT License.