Create bounded box representing the area within the input image that the text was located within?

  • Last Post 05 February 2018 posted this 30 January 2018

I'm still experimenting for just the moment, I do have one quick question.

Suppose I have a 300 pixel by 300 pixel image, and that X:0 Y:0 correlates to the top left corner, and X:299 Y:299 correlates to the bottom right corner.

Suppose I found text in the middle 100x100 px area where the top left corner is at X:100 Y:100 and the bottom right corner is X:199 Y:199.

Is ABBYY capable of giving me a bounded box representing the area within the input image that the text was located within?

I'm working on a GUI testing tool for an interface that cannot be tested any other way. So this particular use case for example would be adding an item to a graphical table, verifying that the item popped up on the screen, and then clicking the item.

So I basically need to add an item with name "Joes Test", then OCR the region where I expect it to be visible, and then be able to click it(so I need to know where it is). Otherwise I have to precapture imagery representing the new item for the purpose of verifying its existence.

Oksana S. posted this 05 February 2018

Hi, hope I understand your question correctly.

There are two processing methods in ABBYY Cloud OCR SDK for recognition of the text fields:

  • processTextField - to extract the value of one text field on an image (no more than 200 characters),

  • processFields - to recognize a lot of small text fields on a page.

For both methods you need to set up some certain recognition parameters which "tell" Cloud OCR SDK where and what kind of text fragment to be recognized. This should be done for all fields, because the automatic layout analysis is not available for field-level recognition. Please refer to the How to Recognize Text Fields article for an overview of possible applications. You can see these methods implementation in our C# GUI Sample. In the sample you select the necessary text field with the mouse, and the program calculates the coordinates of this field, crop it, so that the piece of image contains only needed field and then forms the corresponding request to the service.