ukz-pdf-to-image

Convert PDFs to PNG without losing resolution. Zero-config — works out of the box with a bundled PDF.js renderer. If pdftoppm, pdfimages, or ghostscript are on PATH, they are picked up automatically for higher quality.

Install

npm install ukz-pdf-to-image

Requires Node.js ≥ 20.19.

CLI

ukz-pdf-to-image --pdf input.pdf --imagen output.png              # 300 DPI
ukz-pdf-to-image --pdf input.pdf --imagen output.png -r 3000      # max fidelity
ukz-pdf-to-image --pdf input.pdf --imagen output.png --optimize   # lossless, 4-8x smaller (diagrams)
ukz-pdf-to-image --pdf input.pdf --imagen output.png -f 1 -l 5    # pages 1..5
ukz-pdf-to-image detect                                           # show available backends

Multi-page PDFs produce output-1.png, output-2.png, …

API

import { convert } from 'ukz-pdf-to-image';

const { files, strategy } = await convert('input.pdf', {
  output: './out/page',
  dpi: 300,
});

| Option | Default | Notes | |---|---|---| | output | — | Required. Path prefix for generated files. | | dpi | 300 | Range 36–4800. | | mode | 'raster' | 'raster' renders the page; 'extract' pulls embedded images (requires Poppler). | | firstPage / lastPage | — | 1-indexed, inclusive. | | strategy | auto | Force 'pdfjs', 'pdftoppm', 'pdfimages', or 'ghostscript'. | | optimize | false | Indexed-palette PNG. Lossless on line-art; may band on photos. |

Invalid options throw PdfToImageError with code INVALID_OPTIONS.

Security

Do not run this on untrusted PDFs without process isolation (separate user, container, sandbox). PDF parsers have a long CVE history. Ghostscript is invoked with -dSAFER; pdfjs-dist should be kept up to date (npm audit).

Repository

https://github.com/pauljgl2008/ukz-pdf-to-image

License

MIT