Amharic language ocr

[Bel_Ethio]

Hello there 

I want to develop amharic language OCR,i need some help from the group to help me through the basics!

[Merhawi Fissehaye]

There are no comments from the experts yet, so it wouldn't hurt to share ideas among our selves. I am working on the same topic for my mini-project and I came across the Tesseract-OCR engine. Tesseract is an open-source OCR engine began as a PhD research project in HP Labs, Bristol. It's accuracy outsmarted the other similar commercial engines. In 2005, it was released for open-source. It's trainable and I have found a training data for amharic on the internet contributed by Mr. Sirak to google code: http://code.google.com/p/tesseract-ocr/issues/detail?id=859. I don't think they have made it available for download from the official code repository. You can download it from the link I gave you; but I suggest you read the training procedure first at http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3. It will give you some points as to how the training-data is organized. I am using a python-binding for the tesseract engine called pytesser. I have attached it here.

I hope this helps

[Sree]

Hi Merhawi,

How to  use pytesser along with tesseract ocr engine. What are the tools to be downloaded and what has to be done to make it work. 

[Nick White]

Can you all move this discussion to the tesseract-ocr list (rather
than the tesseract-dev list) please? tesseract-dev is only for the
discussion of development of Tesseract itself. You'll find more
people able and willing to help on tesseract-ocr.

Thanks,

Nick