Thursday, March 6, 2014

Multi-Page Data Extraction

During the past few days, we’ve been working on new features and improvements to data extraction in Athento’s capture module. Last week, we introduced you to the feature that allows users to bring data in from a database, starting from a piece of metadata extracted from the content of the document.  This time, the function that we want to present is that, now, with Athento, we can define metadata to be extracted from documents that are more than a page long.

This feature is applicable to metadata which are extracted using zonal OCR; in other words, for data that we know are always going to appear in the same position.

Let’s take a look at a couple of screen shots of multi-page extraction. In the first image, you can see how the “Application Date” metadata has been defined on the first page of a two-page document:

Next, we can see how, in the same model, we’re already on the second page and we can define the “Immunity from Prosecution” metadata.

We hope that you’ll find this useful! :-) Share

No comments:

Post a Comment