Intelligent OCR Extractor


I’m hoping to use Intelligent OCR activities to extract data from PDF and images. I followed the video in the 2019.4 updates but they use flexicapture extractor in the data extraction scope. I do not have FC currently. Do I need SDK version to create the .fcdot files as input to flexicapture extractor? Would I be better off using the Flexicapture scope of activities, and if so can someone provide a sample?

Alternatively, is there another extractor I can use without needing to get FC? I tried Simple Document Data extractor from Go! and beta version of Machine Learning extractor, but my document is not an invoice or receipt so doesn’t really help.

All help is appreciated!

The best OCR tool for this will depend on the kind of documents you’re working with.
Are they forms or other types of documents?

@Mateus_Cruz What about if its a scan cheque image and extracting only cheque no.

Please note the number printed in last line 101736 is the cheque no.


If all cheque images have the same dimensions, then you can specify that particular region of the image to be the target of the OCR engine.
A similar scenario is discussed here: How to loops list of images and use OCR on them - #5 by Mateus_Cruz