How to train samples in opencv

Question

I'm working on my final year project on ocr in opencv,i have done segmentation of the image and feature extraction,now i have to train to make it recognize the alphabet ,give me an idea of how to do that

Answer 1

I don't know what you have done until now and what feature set you are using. Anyway below links will be a good start.

1) Basic OCR in OpenCV - It uses k-Nearest Neighbours

2) Also try this : Simple Digit Recognition OCR in OpenCV-Python

3) OpenCV samples come with an letter-recongnition sample. Have a look at that!!

Finally, work yourself a lot on your project. Google for your doubts. If you don't find any way, come here, tell what you have done, and now what is stopping you.