Why "convert page to image" and "get the image out of the page" are different jobs
This confuses people constantly because the names sound almost identical: a PDF to JPG converter and an image extractor both produce image files from a PDF, but they're solving completely different problems. Converting a page to JPG rasterizes the whole page — text, layout, background, and any embedded photo — into one new flattened image that looks exactly like the page did. Extracting embedded images instead reaches into the PDF's internal structure and pulls out the original photo file that was placed into the document, at its original quality, with nothing else around it.
If what you actually want is a report's cover photo as a standalone image file — not a screenshot of the whole cover page with headline text included — extraction is the tool for that, and page-to-JPG conversion will give you the wrong result no matter how good the settings look.
Why the extracted file can be byte-for-byte identical to the original
Most photos embedded in PDFs are stored as JPEG data internally, and a proper extractor pulls that data out exactly as it was embedded — not re-compressed, not re-rendered, byte-for-byte identical to the source image. This matters if you need the actual original file for something like a print job or an archive, where even a tiny re-compression pass would be a real quality loss compared to true extraction.
Some image types aren't extractable this way — certain indexed-color palettes and CMYK-encoded images use encodings this kind of tool can't safely convert, so those get reported as skipped rather than silently producing a broken or garbled file.
What to expect when a PDF has several embedded images
A single page commonly contains more than one embedded image object even though it looks like "one picture" to a reader — a photo plus a company logo stamped in the corner, for example, are typically two separate embedded objects even though they appear together on the page. When a document has multiple extractable images across its pages, they come back bundled together in one ZIP file rather than as a dozen separate downloads, which is worth expecting rather than being confused by.