ABBYY FlexiCapture 12 – Receipt Capture Overview

Discover how ABBYY FlexiCapture can use the tools of classification, field extraction, machine learning, and image enhancement to help you capture receipts.

Hello. Today I’d like to show you receipt capture within ABBYY FlexiCapture. Now this is really special technology because it really encompasses the breadth of what we can do within FlexiCapture and all the neat tools that we have. Some of those tools include things like classification, field extraction, machine learning, image enhancement. All of this comes into play when we do a receipt capture project. So on the screen, you can see I have 11 different receipts and the software has extracted an expense type, a total, a vendor and potentially a ton of other things here. I’ll actually show you some of the samples. On the left here is the information that we’re extracting out of the box. Now, of course, just like any FlexiCapture project you can add, remove, modify, but here the software has determined this is a gas station bill, and you can see the other types of information that it’s captured here.

Just kind of continue showing some of these other samples. Here’s a toll bill. Here’s a hotel bill where the software has extracted things like line items as well. Here’s a gasoline bill. Now, sometimes when we look at receipts, it’s actually really helpful that the end user be able to see the original, because remember the technology will enhance the image so that it can read it and extract the best information that it can. But at any point, the end user has the ability to right click and see the original image. And this may be helpful just in context, as a human reading, a document, the original sometimes tells us something that maybe the other image performed by OCR doesn’t tell us. And you can kind of see some of those differences here. You can see the textured background. You can see the lighter text. And those sorts of things, where the software is enhancing that so that we get the best read of the document, but we don’t always sometimes as a human get the full context. So a lot of information here at our disposal.

I’ll just continue showing you some here. Here’s a restaurant bill. Here’s a retail bill. Here’s a parking bill. Toll road. We got a hotel bill as well. And you can see, this is the cool part about the technology is that it’s extracted all of this information literally out of the box and the next thing that applies is the machine learning. So at any point, if I need to teach the software or redirect the software to look for a specific field or fix it, the software will remember those changes so that the next time I process a bill or similar bill, the software will be able to extract that information for us. So that really, really helps receipt capture type of projects become more and more intelligent for us.

Then lastly, I’ll show you here just a car rental. You can see here, it’s not the prettiest image. But you can see here, we got most of that information extracted off of that bill. So receipt capture is really a hard thing for most technologies to do because most technologies really center on only one piece of technology, but ABBYY’s machine learning, image enhancement, classification and field extraction technology combined into one project is really shown very well within this receipt capture type of project. Hope you enjoyed this video. Thank you so much.

, ,

Travis Spangler

Travis writes articles dealing with various technical aspects of document capture and forms processing. He is fluent in Microsoft.NET and holds several certifications including ABBYY FlexiCapture and IRISXTract. As general manager and sales director, he controls the daily operations as well as manages customer accounts to ensure both customers and prospects are receiving the very best from UFC, Inc. Travis has many years of experience with document capture software and content management systems. He also has wide areas of expertise including custom functions in ABBYY FlexiCapture, email API's, Microsoft SQL Server Reporting Services, and many other applications and platforms. He has integrated Amazon Web Services EC2 instances with several applications including the company's CRM system.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.