Answered

Lookup item in output of 'Recognize text in an image or a PDF document'

(0) Share

Report

Posted on by PowerInvader

Hi All,

Would appreciate your help in being able to lookup a particular row that is extracted using 'Recognize text in an image or a PDF document'.

So far I have managed to extract the text from a sample image and create an html table from it.

However, I would like to extract a few particular rows filtered by the X Coordinates and Y Coordinates of the bounding box (please see below table whereby required in yellow is the ExtractedText which should be filtered out by its XCoords and YCoords).

Thank you for your support in advance.

Categories:

AI Builder

I have the same question (0)

All responses (4)

Answers (1)

CedrickB Moderator on at

Like (0)

Report

Bounding boxes are not reliable as it may change from one document to another.

If the text "Please select..." is always the same, you could instead loop into rows and take the text just after to get the value.

Was this reply helpful? Yes No
PowerInvader 19 on at

Like (0)

Report

Yes I am trying to process a standard document where the bounding box remains the same but the Text changes.
Can you please elaborate your solution with some visuals.
Cheers.

Was this reply helpful? Yes No
CedrickB Moderator on at

Like (0)

Report

If the bounding boxes are exactly the same between each document then you can just add a "Condition" action inside the "Apply to each2" to check the values you are looking for where XCoords=TheValue and YCoords=TheValue and do something "If Yes".

But again, that doesn't seems a robust solution.

Was this reply helpful? Yes No
Verified answer

PowerInvader 19 on at

Like (0)

Report

Actually it works really well. Thank you!

Was this reply helpful? Yes No