Get the images out of a PDF
The photo embedded in a report is trapped at full quality — pull the original image data, not a screenshot.
Tools used in this guide
The job
Re-rendering a PDF page captures images at screen resolution with text baked in. Extraction instead finds each embedded image object and saves it as PNG — pixel-for-pixel what was embedded.
Drop and extract
Drop a PDF ≤ 25 MiB / 200 pages. The scanner walks every page's content stream for image objects and lists each with its page and size.
Worked example
A 12-page product sheet with 9 photos yields 9 PNGs — logo at 400×120, photos at their embedded resolution, no rasterized page borders.
Where it stops
≤ 500 images, each ≤ 40 MP; image masks (alpha stencils) are skipped by design. Charts drawn as vectors have no image to extract — render pages with pdf-to-image instead.