OCR PDF

We render each page to an image, run Tesseract OCR over it, and rebuild a PDF with an invisible text layer behind the original image. The page looks identical, but text is now selectable, copyable, and searchable.

OCR accuracy depends on scan quality. Sharp 300 DPI scans of typed text approach 99%; faded photocopies and handwriting are harder.