Well, it's merely an comparison with previously captured images (i.e. images for 0,1,2,3,4,5,6,7,8,9 + point) and translating them to an integer plus decimal.
My advice is for you to try to get somebody with experience in creating optical readers. I have absolutely no problem in assisting someone who wants to help you if him/her is stuck with the coding, but I just can't commit to another new project until I'm clear of what I have on my plate.
Hint: don't pay for it!
There are many documented ways online to capture the screen, i.e.
http://www.codeproject.com/Articles/5051/Various-methods-for-capturing-the-screen#Capture%20It%20the%20GDI%20way
And even more ways to match images, from simple pixel to pixel comparison to SIFT:
http://www.codeproject.com/Articles/5051/Various-methods-for-capturing-the-screen#Capture%20It%20the%20GDI%20way
Another easy approach would be to let a free and proven OCR do all the recognition job. i.e. Tesseract OCR:
https://code.google.com/p/tesseract-ocr/
My advice is for you to try to get somebody with experience in creating optical readers. I have absolutely no problem in assisting someone who wants to help you if him/her is stuck with the coding, but I just can't commit to another new project until I'm clear of what I have on my plate.
Hint: don't pay for it!
There are many documented ways online to capture the screen, i.e.
http://www.codeproject.com/Articles/5051/Various-methods-for-capturing-the-screen#Capture%20It%20the%20GDI%20way
And even more ways to match images, from simple pixel to pixel comparison to SIFT:
http://www.codeproject.com/Articles/5051/Various-methods-for-capturing-the-screen#Capture%20It%20the%20GDI%20way
Another easy approach would be to let a free and proven OCR do all the recognition job. i.e. Tesseract OCR:
https://code.google.com/p/tesseract-ocr/