How to capture fields from the same type of documents but with different patterns?


We have documents that we are using for extraction. We created a FlexiLayout to locate the fields and Document Definition as well. So the twist is that the document format changes. They are not of only one pattern. How to capture fields if their location varies greatly?


If you need to process documents with identical sets of parameters (blocks), identical rules, and written in the same language, but with element positions varying greatly from one document to another, you may create several alternative layouts within your FlexiLayout project.
Each alternative layout corresponds to a particular subset of documents with identically positioned elements. Apart from the layouts, the other processing parameters (pre-recognition/languages, rules, number of pages, etc.) will remain the same for all such documents described by the FlexiLayout. 

Please check the following example:

When matching the FlexiLayout, the program will apply the layout alternatives one by one, top to bottom. If more than one layout alternative matches a document, the first match will be used.

The easiest way to create an alternative within the existing FlexiLayout is by copying it:
1. Click on the "-" next to the SearchElements to hide elements;

2. Holding down the Ctrl key and the left mouse button, drag a copy of the layout alternative to the desired location, e.g. below the current SearchElements group.

3. Upload the document image where the elements are located differently;
4. Locate required fields in the uploaded image and map them using elements from the newly created layout alternative.

Kindly check the following article for more information: Layout alternatives

Have more questions? Submit a request



Please sign in to leave a comment.