08-19-2020 03:01 PM
Hi everyone. I'm trying to create a VI that loads an image and prints the string result. What I have is currently not working when I put the correct training set and image file paths. I've tried including a rectangular ROI wired to the ROI descriptor terminal of the OCR, but even then my result is a dash instead of text. Most online tutorials using OCR to print the string result use a camera to acquire images while I only want to load an image and have it tell me the text on the screen.
Solved! Go to Solution.
08-19-2020 06:21 PM
Try this snippet, first you'll need your train set, you can train you OCR using Vision Assistant, the ROI Descriptor cluster describes the regions of your image to search for text.
08-19-2020 06:29 PM
I forgot to say, this code was made by Vision Assistant.
08-20-2020 09:58 AM
Thanks! I'm trying to replicate this snippet but I'm unfamiliar with some of the functions. Is there a link to this code so I can see what they are called and try it out with my character set file and image?