Word Document Image Extractor

A Python script to extract all images from Microsoft Word documents (.docx) and optionally convert EMF files to PNG format.

Features

• ✅ Extracts all images from .docx files
• ✅ Automatically identifies file types (PNG, JPG, EMF, etc.)
• ✅ Renames files with correct extensions
• ✅ Automatically renames images based on content (using OCR)
• ✅ Attempts to convert EMF files to PNG (when possible)
• ✅ No external dependencies required for basic extraction
• ✅ Command-line interface with options

Installation & Usage with uv (Recommended)

This project uses uv for dependency management.

1. Install uv

# On macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

2. Run the script

You can run the script directly with uv run. It will automatically handle dependencies.

uv run extract_word_images.py o1.docx

3. Project Setup (Optional)

If you want to work on the project or install dependencies in a virtual environment:

uv sync

Manual Installation (Legacy)

If you prefer not to use uv, you can install dependencies manually:

pip install aspose-words pillow

Usage

Basic Usage (with uv)

Extract all images from a Word document, convert EMFs, and rename based on content:

uv run extract_word_images.py document.docx

This will create an extracted_images folder containing all images.

Specify Output Directory

uv run extract_word_images.py document.docx -o my_images

Disable OCR Renaming

If you want to skip the OCR process (faster):

uv run extract_word_images.py document.docx --no-ocr

Extract Without Converting EMF Files

If you just want to extract the files without attempting conversion:

uv run extract_word_images.py document.docx --no-convert

Help

uv run extract_word_images.py --help

How It Works

1. Extraction: Word documents (.docx) are actually ZIP archives. The script:

   • Unzips the document
   • Finds all files in the word/media/ folder
   • Extracts them to the output directory

2. File Type Detection: The script reads the file signature (magic bytes) to identify the actual file type, regardless of the extension.

3. EMF Conversion: The script attempts to convert EMF files to PNG using:

   • Aspose.Words (primary, pure Python)
   • LibreOffice (if installed on macOS)
   • ImageMagick (fallback)
   • Pillow (fallback)

4. OCR Renaming: The script uses EasyOCR to:

   • Read text from each image
   • Generate a descriptive filename (e.g., Biceps_brachii_Site_4.png)
   • Rename the file automatically

Example

$ uv run extract_word_images.py o1.docx
Processing: o1.docx
Output directory: extracted_images
------------------------------------------------------------
Found 9 image(s) in the document.
  Extracted: image8.tmp -> image8.emf
  Extracted: image1.png
  ...

Extracted 9 file(s)

Converting EMF files to PNG...
  ✓ Converted: image2.emf -> image2.png (using Aspose.Words)
  ...

Analyzing images and renaming based on content...
(This requires OCR and may take some time)
  ✓ Renamed: image5.png -> Biceps_brachii_Site_4.png
    (Text: 'Biceps brachii Site 4...')
  ✓ Renamed: image6.png -> Motor_NCS_L_Median.png
    (Text: 'Motor NCS L Median...')

Renamed 9 file(s) based on content

============================================================
✓ Processing complete! Images saved to: extracted_images
============================================================

Troubleshooting

OCR Slowness

OCR is computationally intensive. It may take a few seconds per image. If you have many images and don't need descriptive names, use --no-ocr.

EMF Files Won't Convert

EMF is a Windows-specific format. The script tries multiple methods. If all fail:

1. Ensure aspose-words is installed (uv sync)
2. Install LibreOffice: brew install --cask libreoffice
3. Use an online converter: https://convertio.co/emf-png/

Supported Image Formats

The script can extract and identify:

• PNG, JPEG/JPG, GIF, BMP, TIFF
• EMF (Enhanced Metafile)
• ICO
• And other formats embedded in Word documents

License

Free to use and modify.

Author

Created to simplify image extraction from Word documents.