0

I am working on a webpage to segment handwritten equations and recognize the characters. To do so, my code takes in an image as jpg or png and segments it into seperate images for each character and uses a trained model to recognize them. I would like to add functionality for the user to be able to draw directly in the browser. I’ve been using canvas and created a very basic drawing tool. Is there any easy way to save the drawing as a jpg or png to be able to segment and pass it through the trained model? In principle I am not opposed to having the user be able to download it as png or jpg and upload it themselves but I am looking for an easier way.

Thanks in advance

mmmmo
  • 135
  • 2
  • 6
  • I'm not sure you want a packed image format. AFAIK all model takes raw pixels... – apple apple Mar 11 '19 at 12:01
  • True but the model is already built to unpack user inputed images so I was thinking it would be simpler to have a single input stream. – mmmmo Mar 11 '19 at 12:04

0 Answers0