Optical Character Recognition

Optical Character Recognition (OCR) is a text based recognition and extraction method allowing the user extract handwritten, printed or generated text in images, such as posters, street signs and product labels, or from documents like articles, reports, forms, and invoices. This is typically extracted as words, lines, paragraphs or text blocks. Read more here. OCR is the underlying technique behind Applied AI services such as Azure Form Recognizer which extracts data from documents.

The demo below highlights and returns the extracted text using Azure OCR.


Adjust Font Size and Colour to adjust label for picture.

Please upload an image* with text, in PNG, JPEG, or JPG format. Centeralised single line text works best.

Your Image
Returned Image

Example: User Supplied Image

Example: OCR Returned Image



*All images are stored for 5 minutes and then deleted from the server.

Author: Henry Taylor | Email: henrytaylor@microsoft.com

Copyright: Microsoft.