Convert PDF to PNG with Python, without pdf2image

Choose a local workflow

A script is useful for a repeated task or a folder of documents. For an occasional conversion, the browser tool avoids installing Python packages. Both approaches can keep document processing on your own machine. This example runs on your computer; it is not a hosted API endpoint.

Install the dependency

Create and activate a Python virtual environment, then run python -m pip install pymupdf. Save the example below as pdf_to_png.py and place document.pdf beside it. It uses PyMuPDF directly rather than pdf2image. Consult the official installation guide if no suitable package is available for your system.

from pathlib import Path
import pymupdf

source = Path("document.pdf")
output = Path("output")

with pymupdf.open(source) as document:
    if document.needs_pass:
        raise ValueError("Unlock the PDF before converting it.")
    output.mkdir(exist_ok=False)
    for number, page in enumerate(document, start=1):
        image = page.get_pixmap(dpi=144, alpha=False)
        image.save(output / f"page-{number:03d}.png")

Render and inspect

Run python pdf_to_png.py from that directory. The script creates an output directory, renders each page at 144 DPI and writes separately numbered PNG files. Change the input filename and output directory for a new run to avoid overwriting earlier results. Test a short document first.

Scale carefully

Higher resolution increases memory use. Work page by page, review a sample and retain the source PDF because images do not retain selectable text or interactive features. The example refuses encrypted documents. Review PyMuPDF’s license before incorporating it into distributed software or a hosted service.