[MVP] How to using Project Oxford Vision OCR API with Curl

Category: project oxford


BLShen on Thu, 25 Jun 2015 03:02:35


The Computer Vision APIs are a collection of state-of-the-art image processing algorithms designed to return information based on the visual content, and to generate your ideal thumbnail. The Computer Vision APIs mainly includes analyzing an image, Generating a thumbnail and OCR. The OCR service can help you recognize characters from the photos or pictures.


Build your project for Project Oxford Vision O API with Curl

Environment Initialize

please reference http://curl.haxx.se/docs/sslcerts.html 

and the tools in http://curl.haxx.se/download/curl-7.33.0-win64-ssl-sspi.zip




REM for Basic Authorization use: --user {username}:{password}
REM Specify values for path parameters (shown as {...}), your subscription key and values for query parameters
curl -v -X POST "https://api.projectoxford.ai/vision/v1/ocr?language=unk&detectOrientation=true&subscription-key=XXXXXXXX" -H "Content-Type: application/json" -d "{'Url':'http://images.takungpao.com/2012/1115/20121115073901672.jpg'}" -k


vamsi krishna s on Mon, 28 Mar 2016 13:31:34

Can i use the OCR feature for PDF conversions ?

cthrash99 on Tue, 29 Mar 2016 03:39:12

As noted in the other post, this is not possible today.  You will need some other library that will handle the PDF->image conversion first.

Lee Stott on Tue, 05 Apr 2016 07:00:27

Hi Vamsi

Have you looked at the Windows 10 APIs

Windows 10, UWP and Optical Character Recognition

Windows 10 offers an entirely local, no cloud needed optical character recognition APIs had made it into the Universal Windows Platform. They live in;


If your interested in looking more into this a solution check out my colleague Mike Taulty blog http://mtaulty.com/2016/03/19/windows-10-uwp-a-little-more-ocr/