What is OCR?
OCR, short for optical character recognition, is software that reads the letters in an image of text, such as a scanned page or a photo of a document, and turns them into real text a computer can search, select and copy. Running OCR on a scanned PDF makes it searchable without changing how it looks.
Also called: Optical character recognition, Text recognition
Updated October 14, 2026 · By the PDFZyro team
Why scanned PDFs need OCR
When you scan a paper document, each page is saved as a picture. To you it looks like text, but to the computer it's a grid of pixels, so you can't search it, copy a sentence from it or convert it to an editable Word file. PDFs made from a word processor are different: they already contain real text.
A quick test: try to select a word in the PDF. If you can't, or the whole page highlights as one block, the page is an image and needs OCR.
How OCR works on a PDF
OCR software finds the shapes of letters on each page, groups them into words and lines, and uses the document's language to pick the most likely reading. In a searchable PDF, the recognized text is placed invisibly behind the page image, in the same position as the words you see. The page looks exactly the same, but you can now search, highlight and copy it, and assistive software can read the text.
Recognized text makes a scan useful in other ways too: you can convert it to an editable Word or Excel file, compare its wording with another version, and find it later by searching across all your documents.
What affects accuracy
OCR is very accurate on clean, straight, sharp scans of printed text. Accuracy drops with low resolution, skewed or blurry pages, unusual fonts, complex tables and handwriting. Scanning at around 300 DPI and choosing the document's correct language help most. Always check important numbers and names, because OCR can confuse similar characters, such as 0 and O, or 1 and l.
How to make a scan searchable in PDFZyro
- Open the scanned PDF with Make a scan searchable (OCR).
- Choose the document's language, or up to three if it mixes languages: English, Spanish, French, German, Italian, Portuguese, Dutch or Russian.
- Recognition runs in the background on PDFZyro's own servers. The searchable PDF is saved as a new version and the original scan is kept.