Getting the original pictures out of a PDF
A brochure, a catalogue, a report or a slide deck saved as PDF usually contains photos and graphics that were placed into it. Taking a screenshot gives you a blurry, cropped copy at screen resolution. Converting pages to images gives you the whole page, text and all. What you usually want is the picture itself, at the resolution it was stored in. This tool gets exactly that.
How to extract images from a PDF
- Choose your PDF.
- Pick a size filter. Many PDFs are full of tiny images: bullets, icons, divider lines and logo fragments. “Skip icons and bullets” ignores anything under 32 pixels on its shorter side. “Only large images” keeps only real photos and illustrations (150 pixels and up). Choose “Every image” if you really want everything.
- Click Extract images. Each page is scanned, including images nested inside reusable page parts.
- Browse the previews. Each card shows the page number, the size in pixels, the format and the file size.
- Download single images, or all of them in a ZIP. Names such as “catalogue-p04-02.jpg” tell you the page and the order.
What “original quality” means here
PDFs store images in a few ways, and the tool treats each one as gently as possible:
| How the image is stored | What you get | Quality |
|---|---|---|
| JPEG (most photos) | The JPEG bytes copied out unchanged | Identical to the original, no recompression |
| Lossless (Flate, LZW), RGB, grey, CMYK or palette | Decoded and saved as PNG | Lossless; transparency is kept |
| JPEG 2000 | The .jp2 data copied out unchanged | Identical (needs an app that opens JPEG 2000) |
| JBIG2 or CCITT fax (black-and-white scans) | Counted and reported | Use PDF to PNG for those pages |
Because JPEGs are copied rather than re-encoded, an extracted photo is often sharper and larger than it looks in the PDF. Designers commonly place a 3000-pixel photo and scale it down to fit a column. You get the full 3000 pixels.
CMYK images, used for professional printing, are converted to RGB with a simple formula, so colours may look slightly different from a colour-managed print workflow.
A worked example
A 24-page product catalogue PDF is 18 MB. With “Skip icons and bullets” selected, the tool reports “Found 41 images. 63 small images were skipped by the size filter.” The 41 include 36 product photos (JPEG, about 1600 × 1600 pixels each, copied unchanged), 4 PNG diagrams with transparent backgrounds and the cover image. The ZIP is 15.2 MB, which means almost all of the catalogue’s size was its photos. To make them web-friendly, run them through Compress Image or Resize Image.
Why some pictures are not found
- Vector graphics are not images. Logos, charts and diagrams drawn as lines and shapes, as most charts exported from Excel or PowerPoint are, contain no pixels to extract. Use PDF to PNG to capture the page and crop it.
- Repeated images are stored once. A header logo that appears on every page is usually a single image object reused on each page, so you get it once.
- Inline images (tiny images written directly into a page’s drawing commands) are rare and are not extracted.
Extracting in your browser vs typical online extractors
| This tool | Typical online extractors | |
|---|---|---|
| File uploaded | No | Yes |
| JPEGs kept byte-for-byte | Yes | Often re-encoded |
| Size filter for icons | Yes | Rarely |
| Limits or account | None | Common on free plans |
Respect copyright
Extracting an image does not give you the right to reuse it. Photos in brochures, books and reports usually belong to a photographer or a company. Use extracted images for your own reference, or where you have permission or a licence.