Extract Images from PDF

extract-images.py

# /// script
# requires-python = ">=3.12"
# dependencies = [
#     "pymupdf",
#     "pillow",
# ]
# ///

import fitz  # PyMuPDF
import os
import hashlib
import io
from PIL import Image
import sys

Image.MAX_IMAGE_PIXELS = None  # Disable DecompressionBombError

    
def get_image_hash(image_bytes):
    """Generate a hash of the image to detect duplicates."""
    return hashlib.md5(image_bytes).hexdigest()

def save_image_preserving_quality(image_bytes, image_ext, image_path):
    """
    Saves images in their original quality without recompression,
    unless necessary for format compatibility.
    """
    image = Image.open(io.BytesIO(image_bytes))

    if image_ext in ["jpg", "jpeg"]:
        # Save JPEG as is (no quality loss)
        image.save(image_path, format="JPEG", quality=100)  
    elif image_ext == "png":
        # Save PNG as is (no loss, with optimization)
        image.save(image_path, format="PNG", optimize=True)
    else:
        # Save unknown formats in their best quality available
        image.save(image_path)  

def extract_high_quality_images_from_pdf(pdf_path, output_folder):
    """Extracts embedded images from a PDF with the highest quality possible."""
    doc = fitz.open(pdf_path)
    os.makedirs(output_folder, exist_ok=True)
    seen_hashes = set()
    image_count = 0

    for page_number, page in enumerate(doc, start=1):
        for img_index, img in enumerate(page.get_images(full=True)):
            xref = img[0]
            base_image = doc.extract_image(xref)
            image_bytes = base_image["image"]
            image_ext = base_image["ext"]

            # Generate hash to avoid duplicates
            image_hash = get_image_hash(image_bytes)
            if image_hash in seen_hashes:
                continue  
            seen_hashes.add(image_hash)

            # Save without altering quality
            image_filename = f"page_{page_number}_img_{img_index}.{image_ext}"
            image_path = os.path.join(output_folder, image_filename)

            save_image_preserving_quality(image_bytes, image_ext, image_path)
            image_count += 1
            print(f"Saved: {image_path}")

    print(f"\nExtraction complete! {image_count} images saved in '{output_folder}'.")



if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("Usage: python extract_images.py <PDF_FILE> [OUTPUT_FOLDER]")
        sys.exit(1)

    pdf_file = sys.argv[1]
    # Create default output directory name based on PDF filename
    default_output = os.path.splitext(os.path.basename(pdf_file))[0] + "_images"
    output_dir = sys.argv[2] if len(sys.argv) > 2 else default_output

    extract_high_quality_images_from_pdf(pdf_file, output_dir)