ABBYY FlexiCapture

A complete scanning solution for dynamic data capture, document processing, zone ocr and indexing

The system will significantly reduce the amount of manual data entry required to capture information from your documents, saving you a considerable amount of time and money.

How does it work?

FlexiCapture is supplied as a suite of applications that work together. This allows you to build a bespoke system to process your documents.

The suite includes;

Form Designer - This module lets you create your own paper forms in a way that makes them more suitable for processing with this OCR software.

FlexiCapture - Is the core module of the system. This can be used to create templates, now called Document Definitions, that will process static forms, where the data is always in the same place on each page.

FlexiLayout Studio - This is the intelligent member of the family. If you need to extract data that moves around on the page, or could appear anywhere in your document, you create a flexible document definition. This searches for specific words or phrases in your document to help it identify where the data is that you're looking to extract. Your FlexiLayout is then attached to a Document Definition within the main FlexiCapture application.

This level of intelligence allows FlexiCapture to scan, process and capture data from any zone on any type of document, in any location on the page.

The entire suite can process your documents using a wide range of imaging techniques, such as; optical character recognition (OCR), image character recognition (ICR), optical mark recognition (OMR), barcode recognition, checkmark recognition, document separation, handwriting recognition, document classification, data extraction and document indexing.

This diagram shows an overview of a document workflow. Continue reading about Abbyy FlexiCapture below...

FlexiCapture processes your documents in a very logical way

This scanning and document processing software can be configured in many different ways to cater for the vast array of documents it can process. Some scanning and processing projects can offer a very high degree of automation while others need user intervention. However, the basic process is similar in most cases and this is described below.

Scanning and importing documents into the system

Documents can be brought into FlexiCapture directly from a document scanner that is attached to a PC running the software. They can also be imported from a network folder or an email account.

Importing documents from a network folder can be set up as an automated processing step and the software will monitor the folder and import new documents as they arrive.

The quality of your scanned image files is a key part in achieving accuracy during the OCR recognition process so it is important to ensure that you have the right document scanner.

Matching scanned documents to a document definition

After the documents have been captured (scanned or imported into the system) the software will match them against the appropriate document definition. This ensures that the software knows where to find the data you want to capture off of the document.

Recognition

Depending on how your system has been setup FlexiCapture will run its OCR, ICR, OMR and barcode recognition engines and extract your data.

This will be placed in the correct fields within FlexiCaptures internal database.

Verification - we call this quality control

OCR software has become increasingly powerful and is extremely accurate. However, it has to deal with a wide variety of conditions i.e coffee stains on documents, poorly scanned images, ornate fonts and handwriting.

When the software captures the data you want to extract from a document it is important that it is correct. If the recognition engine is unsure if a character is an 8 or a B it will ask you to verify which is the correct character.

Verification is a quick and easy process because you are only dealing with exceptions (the odd uncertain character). However, it is possible to assist the system in making decisions by specifying the type of data you expect to find in a data field. Using our example above, if you where capturing an invoice total the B would be converted to an 8 automatically because we would set the field to accept numbers but not letters.

Check you business rules

If you have built any business rules into your document definition the system will check that the data captured into each field meets your criteria. For example, you may have created a rule to check if the Net and VAT on an invoice equals the Invoice Total on the document. If it does the system will allow it to pass but if the sum is incorrect the document will be presented to your operator for checking and correction.

Exporting your documents and data

After the data has been captured from your documents and has passed your quality control processes you are ready to export your scanned image files and the associated data.

It is possible to set up multiple export routines to feed different systems. For example, you may wish to create a csv (text file) to feed into your accounts system and simultaneously send the images and an index file to your document management system. This can be used to automatically file away the documents after they have been processed, removing the need to manually index the documents in your filing system.

Image files can be exported in a variety of formats such as Tiff, Multi-page Tiff, PDF, Multi-page PDF, PDF/A, JPEG and BMP etc.

Your data can also be exported in many popular formats such as; csv, xls, xml, to an ODBC database or to Microsoft SharePoint.

During the export setup you can specify where you want your image files saved on your network. If required, the system can automatically build a folder structure for you when the documents are exported.

You have a lot of control over the filenames of your images and data files. For example, you may choose to name invoices by supplier name and invoice number. Alternatively, you may create a folder for each supplier, place the images inside the folder and name them using the invoice number and invoice date.

The options are extensive.

What happens if it goes wrong?

FlexiCapture is an incredibly powerful application and if your project is built correctly it will work extremely well, reducing your document processing costs and saving you many hours labour. Sometimes things do go wrong and a document cannot be processed automatically. The system can be configured to move these files to an exceptions folder for processing by alternative methods.

Different versions of FlexiCapture are available to cater for all needs and budgets

One of the good things about this suite of OCR software is that it is available in different versions.

The product can be purchased as a Standalone system, allowing you to install the software on a single PC. Licensing is very flexible and all licenses provide the functionality described above.

If you need to process thousands of documents every day, or want to split the work up between groups of people, you can step up the ladder and buy the Distributed version of the product. This version of the system can be set up so that some people just scan and assemble documents while others check the quality of your data and correct any business rules to ensure that every document passes your quality control checks before the data is released. This functionality is available in the Standalone version of FlexiCapture but all steps are carried out by one member of staff.

Abbyy supply this application based on the number of images you need to process in a twelve month period and it is easy to move up if your volumes increase.

Please contact us to discuss your data capture requirements.