A PDF keeps its pictures as separate image objects that the pages point to. The tool goes through each page, collects those objects and writes each one out as a JPG, named after the PDF with the page number and a running count, for example report-p4-img7.jpg. Text, tables and drawings made of lines and shapes are not pictures, so they are not exported.

Screenshots of a PDF page capture the picture at whatever zoom your screen shows, plus the page around it. Extracting the image objects directly gives you the picture as it was embedded, often at a higher resolution and without the surrounding text. The Extract Images From PDF page leads to the same tool.

Know what counts as an image. Charts and diagrams drawn in Word, PowerPoint or a design app are often vector shapes, not pictures, and will not appear in the ZIP; render those pages with PDF to JPG or PDF to PNG instead. Pictures with transparent backgrounds come out as JPG, which has no transparency. An image that repeats on every page, like a logo in the header, is saved once for each page it appears on. If the tool reports that no embedded images were found, the PDF contains only text and vector graphics.

Once you have the JPG files, Compress Image to KB shrinks a photo for an upload form, and JPG to PDF puts a selection back into a new PDF. For the words rather than the pictures, use PDF to Text. More export options are on the PDF converter page.

Smaller PDFs are processed in your browser. Larger ones use an encrypted upload to our servers, are handled automatically with no human access, and are deleted within 2 hours. It is free for personal use and needs no account.