How to Extract Images From a PDF Without Software

Two different jobs

Sometimes you want the pages as pictures — a visual copy of each page, including its text and layout. Sometimes you want the photographs that were embedded inside the document, without the surrounding page.

Rendering pages gives you the first. It is what most tools labelled extract images actually do, and it is what you want for sharing a page, putting it in a slide or reading it on a device that handles images better than PDFs.

Rendering pages as images

Open PDF to Image, choose your pages and resolution, and each one comes back as its own file. 150 dpi suits most uses; go higher only if the result will be printed or zoomed into.

This captures the page exactly as it appears, including text, which becomes part of the image rather than remaining selectable.

When you need the embedded originals instead

If you specifically need the source photographs at their original resolution — not a render of the page containing them — that is a different operation, and this tool does not do it.

In practice, rendering at 300 dpi gets you an image at least as detailed as the embedded original in most documents, because the original was usually downsampled when the PDF was made.

Resolution and what you can recover

Rendering a page at 300 dpi produces an image at least as detailed as the embedded original in most documents, because images are usually downsampled when a PDF is created.

Going higher than 300 rarely recovers anything real — it enlarges pixels that were never captured, and produces much larger files for no gain.

Choosing a resolution

Resolution decides both how usable the images are and how large the download is, and the three sensible values are far apart in both respects.

PNG or JPEG

PNG for pages that are mostly text, tables or line drawings. It is lossless, so letter edges stay sharp and there is no fringing around type.

JPEG for pages dominated by photographs, where the file is several times smaller and the difference is invisible. A mixed document usually favours PNG, because smeared text is more noticeable than a slightly larger file.

Selecting just the pages you need

The page range field takes an expression like 1-3, 7, so there is no need to split the document first or render a hundred pages to get four.

This matters more than it sounds on large documents. Rendering is the slow part of the operation, and a phone rendering an entire report at 300 dpi will run out of memory long before it runs out of pages.

Where page images are genuinely the right answer

Slides are the most common case. Presentation software handles images natively and PDFs awkwardly, so a rendered page drops straight onto a slide and stays where you put it.

Sharing a single page is the second. Sending page four as an image is both smaller and clearer than sending a forty-page document with an instruction to scroll. The same applies to anything posted online, since most platforms accept images and none accept PDFs.

Scanned documents

A scanned PDF is already a stack of images, so rendering its pages returns essentially what is inside it. The output is a picture of the page, text included, which cannot be searched.

If you need searchable text from a scan, that requires OCR, which is a separate operation this tool does not perform.

Common questions