OCR
Read the text out of an image or your camera, with the layout kept. Nothing is uploaded.
Pull the text out of a photo, a screenshot, a scan, a whiteboard or your webcam. Keeps paragraphs, columns and headings, reads whitespace-aligned tables, and writes a searchable PDF with the text layered invisibly over the original. Correct a photographed page by dragging its four corners flat, straighten a tilted scan, and threshold locally for uneven light. 27 languages. The recognition engine and every language pack are served from this site, so the image never leaves your browser and nothing about what you are reading goes anywhere.