What Is OCR? How Text Recognition Turns Scans Into Searchable Text
You scan a twenty-page contract, open the PDF, press Ctrl+F to find the termination clause, and the search finds nothing. The words are right there on the screen, yet the computer cannot see them. OCR is the step that fixes this, and once you know what it does, a lot of small document problems make sense.
OCR in one sentence
OCR, short for optical character recognition, is software that looks at a picture of text and works out which letters and words it shows, so the computer can store them as real, searchable text instead of just pixels.
Why a scanned PDF is only a picture
A PDF made from Word or a website contains text: every letter is stored as a character, which is why you can search it and copy a sentence. A scanner or a phone camera works differently. It records how the page looks, dot by dot, and the PDF simply holds that image. To the computer, a scanned letter and a photo of a beach are the same kind of thing.
A quick test tells you which kind you have: try to select a single word with your mouse. If the whole page highlights as one block, or nothing highlights at all, the page is an image and needs OCR.
What OCR does, step by step
The software first cleans the image, for example by turning it into strong black and white. It then finds the lines of text, splits them into words and characters, and compares each shape with the letters of the language you chose, using the surrounding word to settle doubtful cases, such as whether a mark is the letter l or the number 1.
In a searchable PDF the result is added as an invisible text layer behind the scan. The page looks exactly as before, but the text underneath lets you search, select and copy.
What affects accuracy
Clean, straight, printed pages scanned at 300 dpi give the best results. Accuracy drops with blurry or dark photos, pages scanned at an angle, very small print, stamps over the text, unusual fonts, and handwriting, which most OCR handles poorly.
Choosing the right language matters more than people expect. Each language has its own letters and common words, and reading French with English settings misses its accents. If a page is crooked, straighten it first with Deskew PDF.
How to run OCR on a PDF
Open OCR PDF and add your scanned PDF from your device, Google Drive, Dropbox or OneDrive, up to 260 MB. Choose the document's language: English, Arabic, French, German, Spanish, Portuguese or Chinese (Simplified). Press Make searchable and wait; long scans take a minute or more because every page is read. Download the result and test it with Ctrl+F.
If the PDF already contains real text, the tool tells you, and OCR is not needed. For a single screenshot or photo rather than a PDF, Image to Text reads the picture and gives you the text to copy.
What to do after OCR
A searchable PDF is often the first step, not the last. Converting a scan to Word or Excel only works well when the PDF has real text, so run OCR first and then use PDF to Word. If you only need the words, PDF to Text gives you a plain text file.
Always check important details by eye. OCR can misread a digit in an account number or a date, and a single wrong number matters more than a misspelt word.