0

Apologies for the vagueness of my question as I'm just trying to get started with a project.

I would like to have an application that takes a screenshot from my Windows 10 PC, then identify text numbers and specified symbols/images from the the screenshot, and then store these values in a database. I'd like this to be automated from a single click of a command button.

I've stumbled across articles about OCR screen scraping, but I have no idea if this is really possible or how I can go about doing it, or obtaining a software that is customizable to my specific needs.

So basically, I expect the screen have an application open that is updating, and has specific information updating in the same exact locations, so I'd like to program to capture multiple specific target pixel locations and interpret the text characters/digits/symbols/images.

Can anyone help with ideas?

Thanks very much in advance!

gogsy76
  • 1

1 Answers1

0

What your looking for is this project : https://github.com/PaddlePaddle/PaddleOCR

i does the exact work you're looking for which is getting data(symbols-numbers-words..) from an image(screenshot in your case)

the greatest thing about this project is the support of multi-language recognition: Korean, Japanese, German, French, English ...

I think what is left for you is to take a screenshot and parse it to this tool and handle the output.

Best of luck!!!

chikabala
  • 653
  • 6
  • 24