AForge.NET

  :: AForge.NET Framework :: Articles :: Forums ::

OCR neuralne need help

The forum is to discuss topics from different artificial intelligence areas, like neural networks, genetic algorithms, machine learning, etc.

OCR neuralne need help

Postby vattic » Fri May 13, 2011 11:40 pm

hello ,

I try to put all characters i have to the network, and calculate error for each and adjust Weights but i d'ont know The correct way,
i need details for the correct way, Because i try all ideas i have to put all characters and I have bad results .

------------------------------------------------------------------------------------------
Postby andrew.kirillov » Thu May 12, 2011 8:36 am
Hello,

If you train your network providing letters one by one, then it will never work. You need to provide all characters you have to the network. So let’s suppose we speak about training epochs (you will need a lot of them to train network). On each training epoch you provide all 26 characters to network, calculate error for each and average. If error is too big, then start another training epoch, where you provide all characters again.
vattic
 
Posts: 4
Joined: Wed May 11, 2011 9:36 pm

Re: OCR neuralne need help

Postby andrew.kirillov » Sat May 14, 2011 7:29 am

1) Start with samples provided with the AForge.NET Framework (if you use the framework);
2) Take a look at similar projects, like this: Neural Network OCR
With best regards,
Andrew


Interested in supporting AForge.NET Framework?
User avatar
andrew.kirillov
Site Admin, AForge.NET Developer
 
Posts: 3443
Joined: Fri Jan 23, 2009 9:12 am
Location: UK




Return to Artificial Intelligence