Zonal OCR for Single and Batch Document Capture

[marks]

Hello,

I'm new to OpenKM and document management in general. I've been looking for a document management solution that is open source (doesn't necessarily have to be free, it will be used in a commercial environment and we will want to have some kind support contract anyhow). We have some .NET developers on staff and we want to be able to easily integrate our in-house applications with a document management system, and make changes to either as necessary.

I'm looking mainly at OpenKM and Alfresco, but one feature I can't seem to find any information on in OpenKM is Zonal OCR. We want to be able to batch scan documents and have the meta data pre-populated by zonal OCR. Is this possible with OpenKM (even if it's using some other kind of scanning software)? Has anyone done it?

Thanks in advance,
Mark

[jllort]

The idea in OpenKM is using external OCR, scanning and convertir there and then uploading to DMS. There're a lot of OCR engines for example with Omnipage pro can get good results without expensive cost ( has a lot of dictionaries, that one of the keys in any OCR engine ).

There're a lot of OCR engines, from 400€ to several thousands Euros like Ascent Capture. But before suggesting any solution must understand your problem and then we can suggest some option, good in price and good for your needs. Depending which OCR you tries then can be done better or minor integration. For example the most simpliest is connecting OpenKM as network drive, and there uploading files + metadata file. That could be automatically archived by OpenKM with internal processing, etc... More sophisticated could be some minor program that connect with OpenKM via webservices ( java, .net technologies and other ) to synhcronizing OCR output and DMS uploading inputs.

If you want to talk directly with us about professional technical support, please use our contact form at http://www.openkm.com/Contact/ ( we offering supporting services to installation an helping on developing when it's needed ).

[Helliouse]

Hello,
I think marks brings up a very valid point. Zonal OCR is a very important feature. In fact I know we would have a hard time selling Document Management solutions to our customers if they didn't offer Zonal OCR as well as built in OCR capabilities.

I hope the OpenKM looks at Zonal OCR for meta data seriously.

Brad

[jllort]

Probably I've not been clearly on my explanation.

OpenKM can working with any OCR engine. It's not a OpenKM enhancement, feature etc... it'll not be in our roadmap, that's not the idea. The idea is that each client / user has different OCR needs ( features, price limitations, etc... ) in each scenario user must decide with OCR needs and then we can make integrating with OpenKM ( each integration is differents depending OCR capabilities, API if it has etc... ).

The idea is being OCR independant, we don't want to make our own OCR, simply connecting which user has selected.