I want to build a process for extracting a specific tax number from different types of scanned documents.
What is the best approach for extracting the number from documents? the task is a bit complex due to the fact that in each document the number is placed in a different location, with a different label. I’ve used taxonomy to create document types, then digitize document, and classify to appropriate category based on the keyword. Right now only extracting part left, so any ideas, what would be the best option for extraction?
Thanks in advance