Turning PDF pages into images: JPG or PNG?

Rasterising a page throws away everything that makes a PDF useful — selectable text, vector sharpness at any zoom, small file size — so it is worth being clear about when you actually want it.

Good reasons to rasterise

• Embedding a page in a system that only accepts images: a slide deck, a wiki, a chat message, a social post.
• Producing thumbnails or previews for a document list.
• Sending a page to someone whose software cannot open PDFs reliably.
• Creating a visual reference where you specifically do not want the text to be copyable or reflowable.

Poor reasons

• To make the file smaller. It usually will not be; a page of vector text is a few kilobytes as a PDF and hundreds of kilobytes as a 300 DPI image. Use Compress PDF instead.
• To "lock" a document. Rasterising prevents copy-paste but not screenshots, retyping or OCR. It is an inconvenience, not protection.
• To fix a rendering problem. If the PDF displays incorrectly somewhere, the fix is usually in the PDF, not in converting away from it.

JPG or PNG

The two formats compress in fundamentally different ways, and the right choice follows from page content.

JPG uses lossy compression tuned for photographs. It works in blocks and discards detail the eye is least likely to notice. On smooth gradients and photographic texture this is superb: large size reductions with no visible change. On sharp black-on-white edges — which is what text is — it produces "mosquito noise", faint grey speckling around letters, visible at moderate quality settings and ugly at low ones.

PNG is lossless and compresses by finding repetition. A page of black text on white is enormously repetitive, so PNG handles it extremely well: crisp edges and a modest file. On a photograph there is little repetition to exploit, so a PNG can be several times the size of an equivalent-looking JPG.

So:

• Mostly text, tables, diagrams, line art, screenshots → PDF to PNG.
• Mostly photographs, scans of photographs, full-bleed imagery → PDF to JPG.
• Mixed page → PNG if the text must stay crisp, JPG if size dominates.

PNG also supports transparency, which matters if you are compositing a page element over another background. JPG does not.

Choosing a resolution

Resolution is the decision people get wrong most often, in both directions.

A PDF page has physical dimensions, so DPI determines pixel count. An A4 page at 150 DPI is about 1240 × 1754 pixels; at 300 DPI it is 2480 × 3508.

• 72–96 DPI — thumbnails and small previews.
• 150 DPI — the default for anything viewed on a screen. Text is comfortably legible.
• 300 DPI — printing, or feeding the image into OCR. Recognition engines want this much detail.
• 600 DPI — only for reproduction of fine detail, such as archival capture of small print. Four times the pixels of 300 DPI.

Exporting a 40-page report at 600 DPI PNG produces a folder measured in hundreds of megabytes for no benefit if the images are going into a chat thread.

Practical points

Check the first page, then the rest. Documents are rarely uniform; a text report with one photographic appendix may deserve two different exports.

Watch orientation. Rasterising bakes in the page's rotation flag. If a page appears sideways in the PDF, fix it with Rotate PDF first — you cannot un-rotate a wrongly rendered image without re-exporting.

Name outputs in page order with zero-padded numbers, or a folder of 40 images will sort as 1, 10, 11, 2.

Rasterising is one-way. You can put the images back into a PDF with JPG to PDF, but the result is a scan-like document with no text layer. Keep the original.