What works well — and what doesn’t
Text recognition (OCR) is most accurate on clear printed text: book pages, letters, forms, screenshots. Accuracy drops with handwriting, decorative or very small fonts, curved pages, low light and blurry photos. Arabic works best with standard fonts such as those used in newspapers and official letters.
Five ways to improve the result
- Crop the image to the text only — remove margins, photos and backgrounds.
- Photograph the page flat and straight; rotate sideways scans first.
- Use daylight and avoid shadows from your hand or phone.
- Choose the right language: “Arabic” alone is often better than “Arabic + English” for pure Arabic text.
- For a PDF, convert the page to an image first with PDF to JPG.
About privacy and the first run
The first time you use the tool, your browser downloads the recognition engine and the language data (a few megabytes). After that the reading happens on your device; your image is never uploaded. Always proofread the result — numbers and names are worth checking twice.