I am trying to provide an image to a programming language and return a set of data specific to that type of image. The image size may change and location of certain elements may change. However, all elements will be in the same location relative to a specific object.
I will need to extract text and numbers from the image.
This can be done in really any language. I am most confident in my PHP/C++ programming abilities, but I can make anything work. I would probably prefer a Windows app over PHP just for usability.
As far as I know, this should be possible, if complicated. If someone can point me in the right direction I can work from there. A search here and on Google returns nothing more than people trying to parse image links from JSON/XML and such.
Thanks in advance! Comment if you would like any more info
Edit: To be more specific, this is for a friend to help him automate the more tedious parts of leading his Guilds. The images below are two separate games that I would like to be able to use this for. The fields I need to extract are selected. Keep in mind the size of the image may change (they are just screenshots and could be any size)
[image removed]
I will need to extract the image and the corresponding number. The image will need to be matched with a set of images to determine a “name” that the number will go with. Every item on that screen needs to be extracted.
[image removed]
In this game I need to extract the name and both the donation amounts which correspond to that user handle. I will need to do it for every username on that page.
Extracting text from images is called Optical Character Recognition aka OCR. Since you’re mentioning C++ I’d suggest you start with OpenCV.
This looks like a good starting point: Basic OCR in OpenCV
When it comes to extracting images. In that case what you want to start with is probably edge-detection.
OpenCV tutorial including edge-detection