How to copy text from an image
- Choose the language in the image: English, Urdu, or both.
- Drop a photo or screenshot on the mat, or paste one with Ctrl+V.
- Wait a few seconds while the text is read. The first time, the reader and language data download once (about 8–10 MB).
- Check and correct the text, then press Copy text or Download .txt.
Getting the best results
- Sharp and straight: hold the phone parallel to the page, in good light, without shadows. Blur is the main cause of mistakes.
- Crop first: if the picture has a lot besides the text, crop it with Crop image so only the text remains.
- Choose the right language: reading Urdu with the English setting gives nonsense, and the other way round. Use English + Urdu for mixed documents.
- Printed text works best. Handwriting is usually not recognised well.
Urdu OCR
Urdu is written in Nastaliq, which is much harder for computers to read than English because letters join and sit on a sloping line. Cropmat uses the open-source Tesseract Urdu model, which handles clear printed Urdu (books, newspapers, typed documents) reasonably well. Expect to correct some words, especially in small or decorative text.
PDFs and scans
For a scanned PDF, first turn its pages into images with PDF to JPG at High quality, then drop the images here. If the PDF already contains real text, PDF to Word is faster and more accurate.
Everything is read inside your browser; the image is never uploaded.
Questions
Can it read Urdu text from an image?
Yes. Choose Urdu or English + Urdu. Clear printed Urdu works best; check the result for mistakes.
Is my image uploaded?
No. The text reader runs inside your browser, so your image stays on your device.
Can it read handwriting?
Only neat block letters, and not reliably. It is made for printed text and screenshots.