Geek Answers Handbook

OCR in the video? openCV or image processing using OCR?

I need to write a program that makes OCR from a video taken from the screen inside the car before the driver, so it makes OCR only for numbers. I am trying to find a way to implement it. I am thinking of using openCV, but as an alternative, I was thinking of using an OCR program that takes frames from a video and finds numbers. But many OCR programs cannot correctly recognize numbers (perhaps OCR needs training?). So I decided to use a computer vision library to do the job.

What do you think is the best way to implement this simple program?

I'm thin that using a computer vision library with matching patterns would be nice, but maybe OCR can help me. For example, there are programs that recognize license plates.

Therefore, any advice is welcome.

+4
source share
1 answer

You can find a lot on the search schedule.

Checkout this: OCR simple character recognition in OpenCV-Python

It demonstrates simple recognition using OpenCV kNN. The code is in Python, but the functions are standard, so you will not find any problems with porting to other languages.

Some other related links:

  • Main OCR in OpenCV : Code in OpenCV C ++, still using kNN.
  • OCR Signs for Digit Recognition Official : The OpenCV Python samples present have several patterns for OCR digit recognition, and I think one of them is also for video (I have not tried it yet). You can check it out. Link 1 Link 2
+7
source

More articles:


All Articles