How to Extract Text from an Image or Photo with OCR

OCR: When You Need It (and When You Don’t)

OCR (optical character recognition) “reads” text in an image. Use it when the text is in a photo, screenshot, or scanned PDF.

First, try this: if you can select the text in the PDF with your mouse or finger, it already contains real text. In that case, PDF to text is faster and won’t misread any letters.

Step-by-Step: Extract Text from an Image or Screenshot

1. Open Extract text from image.
2. Select the image. On a computer, you can paste a screenshot directly with Ctrl+V.
3. Choose the language: Portuguese, English, or both (useful for technical text).
4. Click Extract text, then copy the result. The first time, the language model (about 1.4 MB) downloads and is saved in your browser.

How to Take Photos for More Accurate OCR

• Shoot straight on: Hold your phone parallel to the page. Angled photos can distort the lines.
• Use even lighting: Avoid shadows from your phone and glare from the flash on glossy paper.
• Frame just the text: The page should fill most of the photo. Crop out the edges and table before processing.
• Adjust the contrast: If the photo is dark or faded, fix it with Adjust brightness and contrast.
• Handwriting is less accurate; the tool works best with printed or typed text.

Make a Scanned PDF Searchable

For entire documents, use PDF OCR. It adds an invisible text layer to each page. The appearance stays the same, but you can search with Ctrl+F, select text, and copy passages. It also works with photos of pages and creates a searchable PDF.

Everything runs in your browser (Tesseract in WebAssembly), so contracts, medical reports, and personal documents aren’t sent anywhere.

Frequently asked questions