I do not know yet if I am going to use ubuntu or Centos on my project...
But I need someone that has experience with training tesseract-ocr.
I need someone that has done this before.... I am going to run tesseract under a binarized image. It is a vehicle license plate image and need tesseract to recognize the plate number.
The image is being prepared to be the best quality possible.
Also I want some suggestions on how to improve it. Maybe a a word ditictionary ?