How to extract images from a PDF
- Drop your file on Drop a PDF here or click to choose. If it has a password, a PDF password box appears.
- Leave What to save on Pictures in the PDF to get the embedded photos, scans and logos. Choose Whole pages to save each page as one image, with a Page resolution of 72, 150 or 300 DPI.
- Pick the File type: PNG (exact) keeps every pixel, JPG (smaller) makes lighter files. Raise Skip pictures with a side under to ignore icons and thin lines.
- Download one picture with its own link, or click Download all as ZIP.
How the pictures are found
A PDF stores each picture once and then draws it on pages. The tool uses Mozilla's open-source pdf.js to walk through every page, collect each image it draws, and decode it to its real pixels. The picture keeps its original width and height, whatever size it appears at on the page. A picture drawn on several pages, such as a logo in a header, is saved once, and its card lists every page.
The pixels are saved as PNG, which is lossless, or as JPG. A small JPEG inside the PDF comes out as a larger PNG with the same look, so choose JPG to keep the size down.
Worked example with the sample file
Click Try a sample 2-page PDF with pictures. The tool reads 2 pages and reports one line: "Found 2 pictures. 1 repeats on several pages and is saved once. Skipped 1 picture with a side under 32 px." The two cards are a 160 x 100 px picture used on pages 1 and 2, and a 96 x 96 px picture on page 2. A 16 x 16 px bullet on page 1 is below the 32 px limit, so it is skipped. Choose Keep everything and the bullet appears as a third card.
Limits and edge cases
- Charts, logos drawn as shapes and text are vector graphics, not pictures, so they do not show up under Pictures in the PDF. Use Whole pages, or the PDF to PNG converter, to capture them.
- A scanned PDF is usually one big picture per page, so you get the scan at its full resolution.
- Saved pictures are re-encoded, so the files do not match the originals byte for byte. Transparent areas are kept in PNG and become white in JPG, and one-color stencil masks are saved as black on white.
- Very large pictures, over about 100 megapixels, are skipped because the browser cannot hold them. If a scan holds words you want to copy, send it to the image to text tool.
PDF image extraction questions
How do I extract images from a PDF?
Open the PDF here, keep Pictures in the PDF selected, and download each picture or click Download all as ZIP. The pictures come out at their original pixel size.
Why does it say no pictures were found?
The PDF probably draws its graphics as shapes and text instead of storing pictures. Choose Whole pages to save each page as an image, or check that Skip pictures with a side under is not set too high.
Do the extracted images keep their original quality?
They keep their original width and height, and PNG keeps every pixel. The pictures are decoded and saved again, so the files are not byte for byte the same as inside the PDF, and a small JPEG can become a larger PNG.
How do I save each PDF page as an image?
Choose Whole pages under What to save, pick a Page resolution of 72, 150 or 300 DPI, then download the pages one by one or as a ZIP.
Can I extract images from a scanned PDF?
Yes. A scan is normally one large picture per page, so you get each scan at full resolution. The words in it are still only pixels, so use an image to text tool if you need to copy them.
Method reference: pdf.js documentation.