Systematic way to program and train tesseract OCR with artificial neural network and c++

[Harsha Jayaweera]

I'm fairly new to tesseract ocr.
I searched everywhere in the internet to find out how to program tesseract with NN. but didn't get any clear idea.

Can someone please tell me/refer me how to code tesseract OCR with NN for training using c++ wrapper clear and understandable way.

If one can please show me simple, trivial example of how to OCR some text with c++(sorry i already know this just asked for clarity and to get intuition) and then extend it to use artificial neural networks programmatically to increase accuracy it would be a great help.

I'm asking this to get a better understanding about how to program using NN. i read few tutorials and documentation a bit but still confused how to do this using c++ and NN.

If someone can please answer this clearly and guide me through this it would be a great help.

Thank You.

[Lavar Askew]

Hi Harsha,

I believe that Tesseract already implements a neural network.  In fact it uses a convolutional neural net.

If you download the source and look in the tesseract/neural_networks/runtime and tesseract/cube directories hopefully that can get you started until someone with a more insightful answer comes along,