Why scanned pages come out crooked
A sheet that slides a few millimetres while the feeder pulls it, a book pressed onto the glass at a slant, or a phone held not quite square to the paper: any of these gives a page that leans. It is still readable, but it looks careless when you send it on, it prints with uneven margins, and text recognition works noticeably worse on slanted lines.
What this tool does to the file
Each page is drawn once in your browser to measure its tilt. The correction itself is written into the PDF as a small turn around the middle of the page. Nothing is redrawn or re-compressed, so the file size barely changes and a page that already had searchable text keeps it. Pages under a twentieth of a degree are treated as straight and not touched at all.
Tips for the best result
- Remove dark borders first. A thick black edge from the scanner lid can pull the measurement off. Trim it with crop PDF and then straighten.
- Mixed documents are fine. Every page gets its own angle, so a stack where only a few sheets went in crooked is fixed page by page.
- Then make it searchable. Straight lines are read far better by text recognition. Run OCR on the straightened file to make it searchable and copyable.
Related tools
- Upside down or sideways pages: rotate PDF turns pages by 90 or 180 degrees.
- Phone photos of documents: scan to PDF cleans shadows and makes the paper white.
- Large scans: compress PDF makes the file small enough to email.