Extract Text from Image

FlexiLayout: How to capture a table using Repeating Group if table header is on each page

In some cases, we might have a table that we are not able to capture correctly using a traditional method – Table element. In such cases, we usually use Repeating Group element.

But what if we come across a multi-page document that has a table header on each page?

We can use two following methods to capture such a table using the Repeating Groups.

Using Absolute search area constraints

To limit the search area to the table area so that it doesn’t capture unnecessary text outside of the table, we can use Absolute search area constraints in the Search Constraints tab.

You can measure the area with the Measure Rectangle tool.

Using nested Repeating groups

Sometimes it might be not suitable to use the Absolute search area constraints method because other tables using this layout might have different positions and lengths of elements, thus making it not convenient to use the method, because you will have to re-measure the area every single time.

In such a case, you can use the nested Repeating group method.

1. Create the first, “main” Repeating group that will include the Table header and footer. 
2. Next, create the nested RG in the first RG. The relations are as follows: 
3. These are the main steps, other elements in the RG don’t need any specific settings and should be designed according to the needed results.

Additional information

FlexiLayout: Capturing a table using Repeating Group

How to reliably capture elements in FlexiLayout Studio if the image resolution can vary

When the image resolution varies, then the search area of elements based on absolute offsets can miss […]

Modern Forms Processing applications have AI-based training algorithms that let users point and click on the location of data in their documents and create OCR templates automatically.

This bypasses the technical requirements of creating complex OCR templates, especially for varied documents like Invoices where the data doesn’t always appear in the same place.

But how good are these AI-based training systems?

In our experience they work well when you have:

• Good quality scanned images
• Clearly labeled data
• Tables with regular columns

Point and click style training doesn’t work quite as well with:

• Poor quality images
• Data that appears within paragraphs
• Tables with overlapping columns, subtotal rows, etc.

These types of documents can still be captured with OCR but they will usually require an experienced technician to manually configure the template.

For natural language data like legal documents, a new artificial intelligence technology called NLP (Natural Language Processing) is available. These work by attempting to “understand” the language used in documents to interpret the location of data points based on meaning. ABBYY FlexiCapture also supports NLP-based training for these types of documents.

Modern Forms Processing software can use rules-based templates for locating data on documents based on label keywords, data types, regular expression pattern matching and other methods.

The most common example in business is an Invoice. Businesses receive invoices from 1000s of different vendors, each with important information like the Invoice Number, Due Date and Total needed to process the document, but each vendor invoice is formatted a little differently than the others.

Software like ABBYY FlexiCapture will look for keywords like “Invoice Number” or variations like “Inv #” and “Invoice No.” to locate the invoice number value on each invoice.

These applications are also able to capture complex table data and output to formats like Excel or a SQL Database, especially when it doesn’t line up into regular columns.

In recent years, artificial intelligence based training has made it possible to simply point and click on the location of data on documents as you process them and generate these templates automatically, dramatically reducing the need for ongoing expert help these systems require.