Hello, Using cloudOCRSDK, I'm using the processBusinessCard method. Can I get the line co ordinates of each word? It looks like this feature is available in processFields method.

I'd not prefer to use processFields because, I want to keep the smartness of processBusinessCards method.

Thanks, Maddy.

asked 09 Jun '14, 21:49

mweb's gravatar image

mweb
113


Hi Maddy,

Unfortunately, processBusinessCards method doesn't assume the definition on the word coordinates. To do that, please, use processTextField method.

link

answered 10 Jun '14, 13:15

SDK_support's gravatar image

SDK_support ♦♦
2763

Thank you. While definition detection need not use the coordinates, this information can be used by our application to clean up some information.

Can you please consider adding the return of coordinates/bounding box of the each field that was detected by processBusinessCards as one of the options? For example:

Sample output XML of processBusinessCards: request to include: left, top, right, bottom <field left=”0” top=”0” right=”199” bottom=”100” type="Name"> <value>John Doe</value> <fieldComponents> <fieldComponent left=”0” top=”0” right=”101” bottom=”100” type="FirstName"> <value>John</value> . . . And also in "Text" field.

If I used processTextField, then I'll loose the benefit of business card detection logic that is present in processBusinessCards. Please consider this option.

link

answered 11 Jun '14, 17:49

mweb's gravatar image

mweb
113

We've sent the request to our analysts if the functionality could be added to the processBusinessCards method. They asked to describe your scenario in details: why do you need to get each field coordinates while recognizing?

(16 Jun '14, 11:06) SDK_support ♦♦

I appreciate the proactive team at ABBYY. There are two reasons: 1. It allows service consumers to add more detection logic. 2. As a software engineering effort, it delinks business card logic and OCR step and allows more control.

ABBYY is great at OCR and analysis. Detecting various types of a business card requires various heuristics that may not be applicable to all users. In such cases, we want to use output of processBusinessCard as a starting point and use location and size information to improve accuracy.

(16 Jun '14, 18:14) mweb

Our analysts are deeply interested in your ideas, but to add new information to method API they need detailed information about the way you are going to use it. So, please, provide us with a detailed scenario of using business card field coordinates: how information about field coordinates could improve the recognition quality? If it's more convenient to describe it in e-mail, please, sent your information to SDK_Support@abbyy.com.

(17 Jun '14, 13:32) SDK_support ♦♦

I need this function too!!! is it possible?

link

answered 06 Feb '15, 11:34

oliver's gravatar image

oliver
111

Unfortunately, there is no such feature in processBusinessCard method yet. You can use processFields method instead. If you need that specific feature in business cards processing, please specify why this feature could improve the results of this method, so that we could make a request for our developers. If it’s more convenient to describe the detailed scenario in email, please contact us at cloudocrsdk@abbyy.com

(03 Mar '15, 11:52) Eugenia Mesh... ♦♦
Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Tags:

×157
×37
×21

Asked: 09 Jun '14, 21:49

Seen: 1,497 times

Last updated: 03 Mar '15, 11:52

© 2016 ABBYY. All rights Reserved. www.ABBYY.com | Privacy Policy | Legal