For example, if document have 4 tags in the form of solid black squares is it possible to determine their coordinates and recognize only text in the area of the rectangle specified by this tags? If so, how this can be achieved? Thanks.
[FR Engine 10] Is it possible to recognize a part of the image by visual tags? Answered
Was this article helpful?
1 out of 1 found this helpful
Comments
1 comment
Is it semi-structured document? Please see ABBYY FlexiCapture Engine, it provides tools for extracting data from documents such as invoices, contracts, or letters which have no fixed layout.
If visual tags have the same location in each document you want to process you can add block with these coordinates manually. Here we discussed how to do this.
Please sign in to leave a comment.