Aspose OCR features

Hi,


I want know about Aspose.OCR feature. What about following questions?

  1. After a document has been ocred does the api provides x
    and y coordinated against each character?
  2. Confidence of words scanned?
  3. Character positioning of characters and their bounds?
  4. Second possible guesses of characters?
  5. Scanning data in tabular format?
  6. Ability to rotate document if required?
  7. Ability to remove noise by applying gusaisan transforms
    and other such functions?

Any help/information appreciated?


This message was posted using Email2Forum by tilal.ahmad.

Hi Masood,


Thanks for your inquiry, We are working over you query. We will update you soon.

Best Regards,

Hi Masood,

Please accept my apology for delayed response. Please find reply to your question as following.

  1. After a document has been ocred does the api provides x and y coordinated against each character?

A rectangle is associated with each character recognized by OCR engine. We can get the coordinates from that rectangle.

  1. Confidence of words scanned?

No, currently we don’t have this opportunity because we don’t have term “word” in our architecture. We have just PartInfo and in most frequent situations it represents word. For every PartInfo we have quality for each symbol and can count in some way confidence of PartInfo.

  1. Character positioning of characters and their bounds?

The same answer as of question no. 1.

  1. Second possible guesses of characters?

Technically we have an internal implementation of this, now we work just with the most plausible symbol. We can think about this feature as a public API for future improvements.

  1. Scanning data in tabular format?

This question is a bit confusing could you please elaborate, if you want to know about an opportunity to return data in formatted style, then unfortunately current API doesn’t not support this feature.

  1. Ability to rotate document if required?

Yes, we can do it by setting rotation skew angle.

  1. Ability to remove noise by applying gusaisan transforms and other such functions?

We use some filters for noise removal, but they are for internal use. We can think about implementing them as a part of public API in future too.

Please feel free to contact us for any further assistance.

Best Regards,

Hi Masood,

In addition to previous reply, I have also logged requested features in our bug tracking system as following. I have linked your request to the issues as well. You will be notified via this forum thread once these features get available.

OCR-33179 – Exposing noise removal functions in public API.

OCR-33180 – Exposing second guess of character functionality in public API.

Please feel free to contact us for any further assistance.

Best Regards,

The issues you have found earlier (filed as OCR-33179;OCR-33180) have been fixed in this update.


This message was posted using Notification2Forum from Downloads module by aspose.notifier.

The issues you have found earlier (filed as ) have been fixed in this Aspose.Words for JasperReports 18.3 update.