Character recognition ANPR

Question

I am implementing the ANPR system. I managed to get a clear area of ​​license plates from the image of a stationary car. I have a problem with certain characters when I feed them with the OCR engine.

I use Tesseract, and the problem is that characters like “W” and “M” are recognized as “N”, and sometimes “U” as “D” and even “O” as “U” or “D” in another way .

I am wondering how to do this with the OCR engine itself, or I can do something to improve the license plate image. Below is the image of the number plate:

I use the Aforge.Net framework with C # and Tesseract as an OCR engine. Any advice would be highly appreciated.

Answer 1

Hi guys, I was able to figure this out by doing the best preprocessing before doing OCR, because my characters seem to be too distorted. it is a pity that I did not close this question :( thanks for your efforts, guys really appreciate this.