Ocr and image pre-processing methods

Question

Hello everyone, I have a collection of images similar to photos of car registration marks.

I am wondering how to pre-process the image before sending it to the OCR engine to determine the text.

I would ultimately want to be able to trim the input only on the white contents of the registration plate.

The actual image will look something like this:

and you should get something like this (color depth is not important)

Thanks for any help. Regards, Bob

Answer 1

• Separate the image (see this ).

• ( ).

• , ( , ). , , - .

• tesseract-ocr, .