Scanned documents and OCR

How scanned PDFs are handled

Lex Cloak runs OCR on every page and works from what is actually visible, because a PDF's built-in text layer can be missing, misaligned, or even hidden. That means scanned PDFs and image-only PDFs are scanned without a separate step, and the engine sees the page the way a person does.

A long scan never leaves you guessing. Lex Cloak shows a checklist of stages (preparing, reading pages, analyzing names, placing redactions, and finishing) and ticks each one off as it completes, next to one overall completion percentage for the whole job. So even on a several-hundred-page document it stays clear which stage is running and how far along the scan is, rather than a single bar that appears to restart at each stage.

When a scanned page detects poorly

• Rotate an upside-down page. A page scanned upside down or sideways often defeats detection. Use the rotate button on the page action float to turn it around, and let it re-scan. Full how-to: rotate an upside-down page.
• Draw over what the scan could not read. For faded handwriting, a stamp partially over text, or a logo, hold Shift and drag to draw a redaction box over it. Full how-to: draw a redaction over anything.
• Redact the whole page on low-quality scans. When a page is too faint or smudged to review match-by-match, press R to redact the entire page.
• Use thumbnails to spot a weak page. The Thumbnails panel shows match counts at a glance, so a page with surprisingly few matches stands out.

Clearer scans detect better than faint or skewed ones. If you have access to a higher-quality source of the same document, scanning that source is usually faster than working around a poor copy. For the background on why scanned files need OCR, see how to redact scanned PDFs. On a Mac, see how to redact a scanned PDF offline on a Mac.