Docly practical guide
How to OCR a Scanned PDF
A PDF from a scanner may contain only page images. OCR recognises characters and adds a text layer, improving search, copying and accessibility.
Steps to OCR a PDF
Add the file to PDF OCR, select its main language and start processing. Download the output and search for a visible word to confirm that the text layer exists.
For phone photos, correct perspective and clean the page with Document Scan first. Skewed, dark or blurred images produce more recognition errors.
Improve OCR accuracy
Use a sharp, straight scan at a sufficient resolution. Select the correct document language because language-specific characters are more likely to be misread under the wrong model.
Review OCR output
Compare names, dates, amounts and identification numbers against the page image. OCR adds convenience but does not replace human review for low-quality scans.
Quick checklist
- ✓Capture pages straight and without shadows.
- ✓Choose the correct document language.
- ✓Manually verify important numbers and names.
Tools in this guide
Frequently asked questions
What does OCR do?
It turns writing inside an image into searchable and selectable text.
Can OCR read handwriting?
OCR is more reliable for printed text; handwriting results depend on the script and image quality.