PRACTICAL GUIDE
How to Convert a PDF to Images on Your Own Machine
Rendering PDF pages to JPG or PNG locally, picking a resolution that stays readable, and being honest about what turning a document into pictures destroys.
Last updated
Converting costs you something — decide it is worth it first
A PDF page is usually instructions: draw this text, in this font, at this position. An image of that page is a grid of coloured dots that happens to look the same. Everything that made it a document goes: the text stops being selectable and searchable, a screen reader can no longer read a word of it, and translation and copy and paste stop working.
It is also frequently larger. A text page weighing forty kilobytes as a PDF becomes several hundred as an image at a resolution anyone can read.
There are good reasons to do it anyway — a system that only accepts pictures, a page to drop into a slide, a preview thumbnail, a form somebody wants as a photograph. Just make sure yours is one of them, because splitting out the pages you need is a better answer when the real goal was a smaller attachment.
Resolution is the entire decision
Seventy-two dots per inch gives you a screen-sized thumbnail: fine for a preview, unreadable if the page has body text.
One hundred and fifty is the sensible default. Text stays comfortably legible on a screen and at ordinary zoom, and the file stays manageable.
Three hundred is print resolution and the right choice only if the result will be printed or if someone will run character recognition over it. It roughly quadruples the pixel count against one hundred and fifty, so expect the files to be large.
Whatever you pick, look at the output at one hundred per cent before you send it. Resolution that seemed generous while choosing a number often does not survive contact with a footnote.
JPG or PNG
JPG for pages that are mostly photographs, where its compression is designed for exactly that and the artefacts hide in the detail. PNG for pages that are mostly text, diagrams, tables or line art, where JPG smears faint halos around the edges of letters and PNG keeps them sharp for a file that is often no bigger. A scanned photograph is a JPG page. A contract is a PNG page.
Doing it without an upload
A browser can render a PDF itself — it is the same machinery that shows you a PDF when you open one in a tab. A page can drive that machinery, draw each page onto a canvas, and hand you the pictures, all inside your own browser with the document never leaving the device.
Off the web there is Preview on macOS, which exports pages as images directly, and pdftoppm from the Poppler tools, which is the reliable command-line answer: pdftoppm -r 150 -png document.pdf page turns a document into page-01.png, page-02.png and so on.
As with any tool that claims to work locally, the test is to load the page, disconnect from the network, and see whether it still works.
Frequently asked questions
What resolution should I choose?
One hundred and fifty dots per inch for anything that will be read on a screen, three hundred if it will be printed or fed to character recognition, seventy-two only for thumbnails. Then open the result at full size and read the smallest text on the page before you trust it.
Will the text still be selectable in the image?
No, and that is the fundamental trade. The text becomes a picture of text. It cannot be selected, searched, copied, translated or read aloud by a screen reader. If any of that matters, keep the PDF and extract the pages you need instead.
Why is the image bigger than the whole PDF was?
Because the PDF stored instructions and the image stores pixels. A page of text is a few kilobytes of drawing commands, and the same page at a readable resolution is a few million coloured dots. This surprises people most with documents that were mostly text.
Is my PDF sent to a server to be rendered?
It does not need to be — browsers can render PDFs themselves, so a page can do the whole job on your machine. Load the tool, turn off the network, and convert. If it still works, nothing was sent.