How to Extract Images from a PDF
A PDF can contain two very different kinds of picture: an image that was placed into the document, and a picture of the whole page. Extracting works differently for each, and knowing which you have explains why the results sometimes surprise people.
The two kinds of picture in a PDF
A placed image is a real image object: a photograph inserted into a report, a logo in a letterhead, a chart exported from a spreadsheet. Extracting these gives you the original file, at the resolution it was placed at.
A scanned page is one large image of an entire page. Extracting it gives you a picture of the page — text, margins, background and all — not the individual pictures that were on the paper.
You can tell which you have by clicking on the picture in a PDF reader. If the whole page highlights as one object, it is a scan.
Extract the images
- Open the Extract Images tool and add your PDF.
- Choose the minimum size. Tiny images under about 100 pixels are almost always decorative bullets and spacer graphics.
- Choose the format: keep the original where possible, or force PNG for line art and diagrams.
- Extract. Multiple images arrive as a ZIP file.
- Extract it and sort by size — the largest files are usually the photographs you want.
Why extracted images sometimes look bad
A PDF stores images at the resolution they were placed at, not the resolution they were created at. If someone put a 300-pixel logo into a document and it was printed at 3 cm wide, extracting gives you those 300 pixels — the page looked sharp because it was small, not because the data was there.
PDF also recompresses on creation. A photograph placed at 90% JPEG quality was re-encoded once already; extracting it does not restore the missing data.
And some PDFs use a mask or an alpha channel. Extracting those can produce an image with a black or white background where the original was transparent.
Common problems
- No images found: the file is a scan. Use PDF to JPG to get the page as an image instead.
- The image has a black background: it used transparency that was lost on export. Try PNG output.
- The image is tiny: that is the resolution that was placed. There is no more detail to recover.
- Wrong colours: the PDF uses CMYK for print. Minor on screen, but expect a shift.
- Hundreds of files: lower the minimum size filter, or use the page range to extract from specific pages.
Getting a better result
If you control the source document, export the image from there rather than from the PDF. The original file has the full-resolution asset.
If you only have the PDF and the page is a scan, extract the page as a high-resolution image and crop what you need. That preserves more detail than trying to isolate a picture inside it.
For a logo, ask the organisation for the brand asset. Extracting a logo from a letterhead gives you a small, recompressed copy that will look soft the moment it is enlarged.
Privacy
Extracting images is often the first step to reusing them — which means the document leaves your control. Think about whether the images contain anything identifying before you share them.
A browser-based extractor reads the PDF and writes the images on your own device. Nothing is uploaded.