Is ABBY OCR available off the Cloud?

  • 2.1K Views
  • Last Post 11 February 2015
tedhohio posted this 10 February 2015

Let me explain better. I am looking for an OCR solution that can be embedded in the server-side of a web application. The idea is to deploy to an organization a complete solution for examining documents and drawings, pulling information (using OCR) from them and using the information in a variety of ways. The users would be interacting using a web browser and the back-end would be RESTful services. The server-side would have an OCR component that the users would access while interacting with PDF document. The nature of the information is proprietary, so the idea of multiple clients interacting with a cloud-based app is impractical. Can ABBYY technology support something like this? I guess I am looking to deploy the core technology on a smaller scale. The SDK would be used to develop the code supporting the interactions with the service.

MFu posted this 11 February 2015

Hi tedhohio,

here some general thoughts on your post:

a) ABBYY OCR technology is available as an SDK for different operating systems and platforms, for often used PC/Server OSes Windows, Linux, OS X there is the "big" SDK FineReader Engine, for Mobile (Android & iOS) there is Mobile OCR Engine and additionally there is a Cloud based OCR service with a RestAPI. Here you can find some comparison documents:

http://www.abbyy-developers.com/en:tech:comparisons:msdk_embedded_fre

http://www.abbyy-developers.com/en:onlineocrsdk:comp_onlinesdk-fre

b) If you license ABBYY technology / SDKs for your own backed, you have to integrate it via code and the API provided by the toolkit. This is also possible with scripting languages.

http://www.abbyy-developers.com/en:tech:samples:scripting_languages

c) The ABBYY Cloud OCR SDK comes with a REST-API as you probably have seen but if you want one you can use with the other ABBYY tool-kits you have to develop them by yourself.

d) not sure if I got that right, but if you only want the user to process some zones on the image/documents they have in the browser, you could test the zonal/field-level call from Cloud OCR http://ocrsdk.com/documentation/apireference/processFields/

e) What I described is targeted to developers that create a solution/application. ABBYY also offers ready to use products like Recognition Server that can be integrated with a SOAP-API. But this approach needs much less coding, but you also have less granular control over the processing steps.

If you are interested to test one of the ABBYY tool-kits you should get in contact with the ABBYY Office/Partner in your region. The easiest way to do this is via the contact form on http://www.abbyy.com/ocr_sdk/

Hope it helps - with regards

Michael (working at ABBYY)

Close