How to Extract Images from a PDF (Original Quality)
Pull the actual embedded JPG and PNG files out of a PDF at original resolution — no screenshots, no quality loss. Free, no sign-up, 30 seconds.
Someone sent you a PDF — a brochure, a report, a product sheet — and you need the photos out of it. The instinct is to zoom in and take screenshots. Don't. The way to extract images from a PDF at full quality is to pull out the actual embedded files, because the PDF is carrying the originals inside it. A free browser tool does this in about 30 seconds and returns the images exactly as they were placed in the document.
Why screenshots lose quality (and extraction doesn't)
A PDF doesn't "draw" its photos from scratch — it embeds the original image file (usually a JPEG or PNG) and an instruction saying "place this image here at this size." A photo placed in a print-ready brochure is typically stored at 300 dpi; a 3×4-inch photo at 300 dpi is roughly 900×1200 pixels of real data sitting inside the file.
When you screenshot that photo on screen, you capture it at whatever size it happens to render in your viewer — usually the equivalent of 72-150 dpi. On top of that, the pixels have been resampled once by the PDF renderer and get recompressed again when you save the screenshot.
| Method | What you get | Quality |
|---|---|---|
| Screenshot the page | 72-150 dpi re-render, recompressed | Soft, small, artifacts on text and edges |
| Copy-paste from viewer | Viewer-dependent, often downscaled | Unpredictable — sometimes fine, often not |
| Extract embedded images | The original stored file, byte-for-byte | Identical to what the author placed — often 300+ dpi |
Extraction is a copy, not a capture. If the author placed a 4000×3000 photo, you get a 4000×3000 photo — even if it's displayed the size of a postage stamp on the page.
Extract every image in 30 seconds
Open the tool
Go to Extract Images in any browser — no install, no account.
Upload the PDF
Drag it in. Free up to 20 MB and 50 pages, which covers most brochures, decks and reports.
Extraction runs automatically
The tool scans every page for embedded image objects and copies each one out untouched — 2-10 seconds for a typical file.
Download the ZIP
All images in one archive, named by page and position (e.g.
page-03-img-01.jpg) so you can trace each image back to where
it appeared.
Done
No watermarks. Your upload is auto-deleted from our server within 1 hour.
What you actually get back
- Formats. Images come out in the format they were stored in — photos are almost always JPG, graphics with transparency or sharp edges are PNG. No conversion, no recompression.
- Resolution. Whatever was embedded. Print-quality PDFs often carry 300 dpi images; web-exported PDFs may only carry 96-150 dpi versions. Extraction gives you the best copy that exists in the file — it can't invent pixels that were never there.
- Naming. Files are numbered by page and order of appearance, so a 20-page catalog with 60 photos stays navigable.
- What's skipped. Vector logos, charts and line art aren't images at all — they're drawing instructions, so there's no file to extract. Tiny decorative fragments (bullets, borders) may appear in the output; just ignore them.
Scanned PDFs: one big image per page
If your PDF came from a scanner or a phone scan app, every page is a single full-page image — there's no separate "photo" object to pull out. Extraction still works perfectly: you'll get one image per page at the scan's full resolution.
Decide what you actually want first:
- The whole page as an image (to re-insert somewhere, or archive) — extraction gives you exactly that.
- A photo that appears on the scanned page — extract the page, then crop the photo region in any image editor. The tool can't separate a photo from the paper around it, because to the PDF it's all one picture.
- The text on the scan — that's not an image job at all; run OCR instead. See the best free OCR for PDF.
Only need images from a few pages?
For a 200-page report where the photos you want live on pages 12-18, extract those pages first with Split PDF (the range workflow is covered in extract pages from a PDF free), then run the smaller file through Extract Images. Faster upload, smaller ZIP, no digging through hundreds of files. The free tier's 10 operations per day easily covers a two-step job like this.
Frequently asked questions
Will the extracted images match the original quality?
Yes — exactly. The embedded image file is copied out byte-for-byte, with no re-rendering and no recompression. What the author placed in the PDF is what you download.
What formats do the images come out in?
The format they were stored in — JPG for most photos, PNG for graphics with transparency. Some PDFs use less common internal encodings (like JPEG 2000 or CCITT fax for scans); those are converted to a standard format losslessly so every file opens anywhere.
Why are some extracted images low resolution?
Because that's all the PDF contains. Tools that "export to web" often downsample images to 96-150 dpi to shrink the file. Extraction returns the best version that exists inside the PDF — it can't upscale what was never embedded.
Why didn't a logo or chart get extracted?
It's probably vector art — lines and shapes drawn by instructions, not a stored image file. There's nothing to extract. If you need it as an image, screenshot that element at high zoom (400%+) as a workaround.
Can I extract images from a password-protected PDF?
You need the open password. Unlock the PDF first, then extract. No legitimate tool bypasses a password you don't have.
Some images come out inverted, masked or in pieces — why?
PDFs sometimes store an image plus a separate transparency mask, or slice a large image into tiles. Most extractions are clean, but a heavily optimized PDF can produce a few odd fragments. Take the good files, ignore the rest.
Is there a limit on the free plan?
Free handles files up to 20 MB and 50 pages, 10 operations per day, with no watermarks and no sign-up. Pro is $9/mo for unlimited files, pages and operations.
Is it safe to upload a confidential PDF?
Files are processed for the extraction and auto-deleted from the server within 1 hour — nothing is kept, and outputs carry no watermarks. For material you can't upload at all, use a local desktop tool instead.
The pdfty team builds privacy-first online PDF tools — compress, convert, OCR, sign and protect. Files are deleted within 1 hour. About us →
