We use Tesseract.js, a powerful open-source OCR engine that runs entirely in your browser. This means text extraction happens locally on your device, without any data leaving your computer.
To improve accuracy, images are processed using a WebAssembly build of OpenCV, a leading computer vision library. This allows for fast and efficient image cleaning, resizing and filtering — all done directly on your machine.
Your images never get sent to our servers. From start to finish, the entire recognition process takes place in your own browser window. Once the page is fully loaded, Text Scanner can continue working even without an internet connection. Perfect for low-connectivity environments or users who need guaranteed offline privacy.