OCR PDF

OCR a Scanned PDF or Image Right in Your Browser

A scanned document is really just a picture of a page. You cannot select, copy, or search the words inside it, because there is no text layer — only pixels. OCR (optical character recognition) solves this by reading the shapes of the letters and turning them back into real, machine-readable text. PDFMarkr runs OCR entirely in your browser, so you can convert a scanned PDF to text or pull image to text without uploading anything to a server.

How the OCR tool works

PDFMarkr uses tesseract.js, a well-established open-source OCR engine, running locally on your device. You open a scanned PDF or an image, the tool reads each page, and it gives you the recognized text. It can also rebuild your file as a searchable PDF, adding an invisible text layer over the original scan so the pages look identical but are now selectable and searchable.

The first time you run OCR, your browser downloads a roughly 11MB English language model. That download happens once and is cached afterwards, so later runs start much faster. Because everything is processed in memory on your own machine, your document is never sent to us or anyone else.

Common use cases

• Making an old scanned contract or report searchable so you can find a clause by keyword.
• Copying text out of a photographed receipt, form, or book page.
• Turning a scanned invoice into text you can paste into a spreadsheet or email.
• Preparing an image-only PDF so its contents can be indexed and quoted.

Extract text vs. OCR — which do you need?

If your PDF was created digitally (exported from Word, a browser, or design software), it usually already contains a real text layer. In that case you do not need OCR at all — the faster route is our Extract Text tool, which pulls the existing text directly. OCR is specifically for extracting text from scanned PDFs and images that have no text layer to begin with. Once your document is recognized, you can also send it to our PDF to Word converter to keep editing it.

Honest limitations

OCR is powerful but not perfect, and we would rather be upfront about it. This tool currently supports English only. Accuracy depends heavily on scan quality: a clean, high-resolution, straight page reads well, while blurry, skewed, low-contrast, or noisy scans produce more mistakes. It is built for printed or typed text and is not designed for handwriting. For any important document, always proofread the recognized text before relying on it. If your source pages are crooked or upside down, straightening or rotating them first will noticeably improve results.