I am trying to use OCR API for some document with Table Structure… its working fine i am also getting output of the OCR in Json… with TOP X , Left X etc now i want to store same data in database with Field Name / Value example for Item Name / Price from database how can we identify Header / Value etc…
So you want to extract certain infos from the OCR’ed text? The standard way to do is to use regular expressions.
Examples:
The other option is to use the coordinates of the word bounding boxes. This works if you know that certain data is always at a certain position, e. g. when scanning always the same type of invoices.