OCR
OCR (optical character recognition) finds the text in scanned or photographed pages and adds it to the PDF, so you can search, select and copy it.
Run OCR#
- Select one or more PDFs and open OCR / Cleanup scans.
- Under Languages, choose every language that appears in the document. At least one is required.
- Pick an OCR Mode and any Processing Options (below).
- Click Process OCR and Review, check the result, and download it.
Options#
OCR Mode
- Auto (skip text layers) - the default. Skips pages that already have text.
- Force (re-OCR all, replace text) - OCRs every page and replaces any existing text.
- Strict (abort if text found) - stops if any page already has selectable text.
Processing Options
| Option | What it does |
|---|---|
| Compatibility Mode | Makes larger files that work better with some languages and older PDF software. |
| Create a text file | Also saves the recognised text as a separate .txt file. |
| Deskew pages | Straightens tilted pages before recognition. |
| Clean input file | Removes noise and boosts contrast before recognition. |
| Clean final output | Tidies the text layer in the finished PDF. |
In the desktop app, OCR usually needs Stirling Cloud or a server (Where your tools run). On a self-hosted server, if a language is missing, ask your admin (OCR languages).
