Extract images from PDF files without losing quality
The product photos you need are stuck inside a supplier catalog, or the only copy of your logo is in an old brochure PDF. Screenshots look blurry. Your todo.is agent pulls the images out of the file at the resolution they were stored, skips tiny icons if you want, and names each one by page.
The prompt
- Extract the images from the attached PDF: [ATTACH THE PDF]. Take images from [WHICH PAGES]. Save them as [FORMAT] at their original resolution, and skip anything smaller than [MINIMUM SIZE]. If an image is a vector drawing (logo, chart) rather than a photo, render that page area as a high-resolution PNG instead. Name files by page number and order, put them in a ZIP and give me a list with each image's size in pixels.
What to change
- [ATTACH THE PDF]: Attach the PDF (up to 50 MB).
- [WHICH PAGES]: E.g. "all pages", "pages 4-20" or "only the product pages".
- [FORMAT]: E.g. "original format", "PNG" (best for logos and screenshots) or "JPG" (smaller, for photos).
- [MINIMUM SIZE]: E.g. "200 pixels wide" to skip icons and bullets, or "nothing, keep all".
Example result
- Image extraction: Spring_Catalog_2026.pdf
- 36 pages · 51 images saved · ZIP 64 MB
- What is in the ZIP
- • p01_01.jpg · cover photo · 3000 x 2000 px
- • p04_01.jpg to p04_04.jpg · product photos, garden chairs · 1600 x 1600 px each
- • p05_01.png · product on transparent background · 1200 x 1200 px
- • ... 44 more, listed in Images_list.csv (file, page, width, height, format)
- Logos and charts (vector)
- • The brand logo on page 1 and two charts on page 30 are drawn as vectors, not stored as pictures
- • Your agent rendered those areas as PNG at 600 dpi, so they stay sharp. Files: p01_logo.png, p30_chart1.png, p30_chart2.png
- What was skipped
- • 112 tiny images under 200 px wide: bullet icons, social media icons and decorative lines
- • A repeated background pattern used on every page (saved once as p00_background.jpg in case you want it)
- Notes on quality
- • Images are saved exactly as stored in the PDF. If the designer placed a small, low-resolution picture, that is all the PDF contains
- • Some photos were stored as JPG with CMYK colors for print; your agent converted them to RGB so they look right on screen
- • Image p12_02 had a soft mask (transparency). It is saved as PNG to keep the transparent edges
- Usage reminder
- • Images in a catalog or brochure usually belong to the company that made it. Check you have permission before reusing them
How to do it with todo.is
- Copy the prompt and choose the pages and image format.
- Paste it in todo.is on the Today screen with the PDF attached, or send it to your agent on WhatsApp, Telegram or email.
- Your agent pulls out every image, renders vector logos and charts sharply, and zips the files.
- Download the ZIP, or reply "only the product photos, as JPG" to narrow it down.
Tips for a better result
- Extracting gives you the image as stored in the PDF, which is sharper than any screenshot. You cannot get more pixels than the original though.
- Logos and charts are often vectors. Ask for them as high-resolution PNG, or as SVG if you need to edit them in a design tool.
- Set a minimum size to skip icons. A catalog can contain hundreds of tiny decorations.
- Need a white or transparent background on the extracted product photos? Ask your agent to remove the background in the same to-do.
extract images from PDF: FAQ
- Can I extract images from a PDF without losing quality? Yes. Your agent saves each image at the exact resolution stored in the PDF, without recompressing it unless you ask for JPG.
- Why are some images missing? Logos, charts and drawings are often vectors, not pictures. Your agent renders those page areas as high-resolution PNG files instead.
- Can I get images from a scanned PDF? A scanned PDF is one big picture per page. Your agent can save each page as an image, or crop out the photos on it.
- Can I extract only the images from certain pages? Yes. Give page numbers or ranges, like "pages 4 to 12", and only those images are saved.
JavaScript is required to use the todo.is app.