I'm looking for some supervised machine learning techniques for pattern recognition on image.
Currently I'm using EmguCV for detecting lines and as result pattern, but it's not enough accurate for me,
and I don'n know how can I make implementation for selection some particular area ROI and use it as instance for training
for looking similar objects on other images then store it at CSV table as wage coefficients and use it for neurons and dendrites.
And with increasing database (supervised training) I will be able to get better recognition rate for each pattern type on image.
So it's not very trivial task, but I will be very thankful for any help, especially with code examples.
Preferred .NET (C#, VB.NET) that I can use AForge.NET, Accord.NET of EmguCV.
THANKS!!!