I am exploring ABBYY for my project use case. We have Invoices, Cheques, and few other document types to be classified and extracted. I was going through ABBYY FineReader and FlexiCapture capabilities and could not come across classification feature for ABBYY FineReader.
Does ABBYY FineReader have classification capability? If so, Does ABBYY FineReader trial version offer this capability? How can classification be done using FineReader?
Much appreciate your help. Thank you!
You can setup classification for both FineReader Engine and Flexicapture Engine. From my own experience, your use case is pretty common and ABBYY developped many specific tools in order to support these use cases as many enterprises need automation workflows based on invoices.
The trial version of both products should support classification by default, expecially with FlexiCapture as it is one of the main goals of this SDK. If it's not the case, you may send an e-mail to your sales contact at ABBYY to obtain trial licenses with that option.
Here is a link to ABBYY portal with some information about FRE vs FCE to help you choose the ABBYY product based on your needs, and a pretty exhaustive guide for setting up classification in FRE and FCE, with code samples.
Good luck !
To classify documents with Abbyy FineReader or FlexiCapture, The document types should be predefined. From your post, It is clear the document types are cheques, invoices etc.,
Both FineReader SDK and FlexiCapture provide classification as part of developer trial or full licences.
But FlexiCapture will be preferable if your document structure is standard. Both the tools provide Classification capabilities such as classifications based on Bar code, Author, Images etc., These classifications will require your own algorithms.
FRE is SDK so all the business and FRE logics are to be written by you but Flexicapture has UI capabilities. You could also try FlexiLayout from Abbyy. FlexiLayout templates can be passed as argument to Flexicapture to recognition and extraction after the classification.
As the owner and power-integrator of both FineReader Engine SDK and FlexiCapture line of products, I say that FineReader has "simple classification" and will lack capabilities for post-classification data extraction on document types like Checks and Invoices (unless you process only a few formats), which are highly variable and require either unstructured data capture or machine learning technologies, which are not in FineReader.
Even ABBYY describes the differences in classification here on their website.
Each tool has its purpose and intended use. I have implemented Invoice (more info) and Check to processing with integrated Check21 capability (more info) using ABBYY FlexiCapture. I would not do it with FineReader, hard to achieve wide scalability across formats (although possible with heavy text parsing) and likely will quickly reach limits (managing too many templates and strucutres).
Even ABBYY themselves released "FlexiCapture for Invoices" product and there is no FineReader for Invoices, so ABBYY clearly state which product should be used.
