tags:

views:

1822

answers:

6
+13  Q: 

Open source OCR

Is there any open source OCR library written in .NET, or written in any language but can be used in an ASP.NET application?

+1  A: 

Tesseract OCR was developed by HP and open sourced, I believe. I'm not sure how good the quality is.

paxdiablo
+2  A: 

There is Tesseract and although it has no native .net bindings, it's pretty complete.

The "one to watch" is OCRopus - A Google-sponsored effort that isn't ready for production work yet.

Oli
OCROpus can use the Tesseract engine, and does some layout analysis and other image pre-processing.
sventech
+1  A: 

For what it is worth, there is also GOCR. Not tried.

PhiLho
+7  A: 

Use Tessnet.

Tessnet is C++/CLI .NET Wrapper for tessdll (and tesseract) for .NET 2.0.

SelvirK
+1  A: 

As the others are writing Tesseract is a good open source OCR implementation.

Eventhough it is good it is only an OCR engine. It does not do layout detection. If your input is not pure text you could have to figure out the layout yourself.

Morten Holdflod Møller