T O P I C R E V I E W |
indyrhett |
Posted - Jul 21 2018 : 12:31:19 Has anyone had any luck reading Numbers with the OCR that are segmented like a digital display? I'm not having a lot of luck at the moment.

Thanks,
Rhett Price IndySoft
|
4 L A T E S T R E P L I E S (Newest First) |
xequte |
Posted - Jul 22 2018 : 18:58:41 Hi Rhett
It looks like you should be able to do the same with a set of 14 segment fonts.
You can download the Serak trainer from:
https://github.com/serak/serak-tesseract-trainer
Nigel Xequte Software www.imageen.com
|
indyrhett |
Posted - Jul 22 2018 : 16:40:28 Thanks for the help.
This was the example video for the 7 segment.
https://www.youtube.com/watch?v=i_1-hGsXxy8
Hmm.. interesting. I don't have a lot of experience on isolating each number like that but if I can't get OCR to do it might have to go that route.
Thanks,
Rhett |
xequte |
Posted - Jul 22 2018 : 16:35:06 Hi Rhett
If you have seen it work with Tesseract, then it should be possible with IEVision. Are you able to find any 14 segment Tesseract training sets?
Of course, you should also consider the other possibility of passing each character image to a method that checks on/off state of each of the 14 segments to determine the character.
Nigel Xequte Software www.imageen.com
|
indyrhett |
Posted - Jul 21 2018 : 13:00:50 To further clarify, I believe this would be a 14 segment display (appears there are 7, 14, and 16 segment displays around)
I found some users using Tesseract OCR freeware doing it - but would much rather figure out with ImageEn.
Thanks,
Rhett
|