SwiftOCR icon indicating copy to clipboard operation
SwiftOCR copied to clipboard

Published 20 hours ago • verified publisher icon NMAC427

Training Images

Open acegreen opened this issue 6 years ago • 0 comments

Hello, thanks for the great work on this library.

I have spent some time trying to understand how to train a new network with a set of images. I noticed you have a set of training images that are just of the background. I would like to use the same images I used to create my ML model

Below is a sample of the image I would like to use. I have used CoreML and ARKit to detect the box and would now like to read two set of texts

"SN: OM+8000002274" "PIN: 719810"

What would you recommend I do to detect these two texts?

acegreen avatar Feb 16 '18 12:02 acegreen