Extract the images from a PDF
The images inside a PDF, pulled out one by one into their own file.
Your files never leave your deviceDrop your PDF here
Images that are actually embedded are detected automatically
How to extract images from a PDF, step by step
Extracting images recovers the images actually embedded in a PDF, each into its own file — not the pages themselves.
Drop in the PDF to analyze
Drag the file onto the drop area, or click to browse for it. VellumPDF scans every page looking for embedded images.
Let the analysis finish
A progress bar shows the current page while the document is being examined, before the number of images found is displayed.
Browse the images found
Each image appears with its dimensions in pixels and its own individual download button.
Download all of them or just one
One button downloads every image at once as a ZIP archive; each image also has its own individual download button.
An image smaller than 16 pixels on a side is ignored: it is almost always a layout artifact (a stretched gradient, a rasterized bullet point) rather than a genuine image worth extracting. If no image is found, the document either contains only text, or its pages are themselves plain images — a scan, for example — in which case the “PDF to images” tool is the right one to use.
Frequently asked questions
What is the difference between extracting images and converting a PDF to images?
Extracting images pulls out only the images already embedded in the PDF — photos, logos, illustrations — as they were inserted. Converting the PDF to images, by contrast, turns each whole page, text included, into a rendered image.
What format are the extracted images saved in?
Every recovered image is exported as a PNG file, whatever its original format inside the document. All the images extracted from a given PDF are bundled into a downloadable ZIP archive.
Is this tool free, and does the PDF stay on my device?
Yes, extraction is free, with no limit on the number of files, and it runs entirely in your browser. The PDF is never sent to a server.
Does a logo on every page come out a hundred times?
No, the tool de-duplicates: an image reused from page to page is exported once. De-duplication relies on the object’s identity inside the file; the rare images that carry none may come out more than once.
At what quality are the images exported?
As they are stored inside the PDF, with no lossy re-encoding: the tool does not invent a resolution that is not there, and does not degrade what is. Every image is exported as PNG, regardless of its size.
What happens if an image refuses to extract?
It is simply left out of the results, with no blocking error message: some rare internal formats, or images generated on the fly, cannot be pulled out cleanly. If the number of images found seems too low, converting the whole page to an image is the fallback.