(OCR ) How to Recognise Handwritten fractional numbers using Neural networks

Question

I want to be able to recognise handwritten math numbers using images of the numbers , i was able to do create a ANN model for recognising simple decimal numbers , but i have no idea on how to recognise fractional number. What kind of ML model do i need to recognise fractional numbers (p/q forms). As i am relatively new to machine learning, i have no idea how to do it. Any study material will be appreciated.

Answer 1

Usually one of the steps is "segmentation", where we separate the image into smaller images, one for each character/symbol. Thus the fraction "1/13" would have four character, "1", "/", "1", and "3". Then, you apply a neural network separately to each of those four smaller images.

See https://en.wikipedia.org/wiki/Optical_character_recognition for an introduction. OCR is covered in many standard sources. I suggest reading the literature on the subject to learn more about standard methods.