PdfRecast guide
Edit image-based PDF pages with OCR-assisted text boxes
A scanner normally stores a page as pixels rather than editable letters. PdfRecast uses optical character recognition to locate likely text, places reviewable boxes over the scan, and lets you correct recognized content before saving a new PDF.
How to use this workflow
- Open the scanned PDF and render the page clearly at the correct rotation.
- Run text extraction and wait for OCR to identify lines and words.
- Compare each important OCR box with the page image and correct recognition errors.
- Save a new copy and verify that additions align correctly at normal and high zoom.
How to improve OCR accuracy
Clear, straight, high-contrast scans work best. Shadows, folds, stamps, handwriting, low resolution, mixed languages, decorative fonts, and text over patterned backgrounds make recognition harder.
OCR output should be treated as a draft. Carefully check identity details, dates, amounts, reference numbers, and any text that will be used for an official decision.