Removing unidentifiable characters from manual ICR
First of all I want to say thank you for providing queXF, it's a great suite!
I wonder how to get of rid of unidentifiable characters when doing manual training.
Sometimes people write two characters into one box when they don't respect the max length.
There is no correct letter for these cases, so i click them and they get red. However, they stay there when I do the next manual ICR and I am afraid to perform automatic ICR in a state like this (the rationale behind this seems to be to keep it until the correct letter is entered beneath the box - is this correct?)
I diff'ed the DB before and after accepting the letter as some random char, but apart from increases in ocrkb, ocrkbdata, and ocrtrain I could not find where the actual info whether a glyph needs training is stored.
As an intermediate solution I'd be glad to know where to drop these glyphs in the DB.
In the long run, it would be practical to either have some icon on the glyphs to completely remove them or some function to remove all unidentified, remaining glyphs. Or probably just another button next to "Train": "Train and remove red chars" or something along these lines.
I think this is https:/
Thanks a lot in advance!
Question information
- Language:
- English Edit question
- Status:
- Open
- For:
- queXF Edit question
- Assignee:
- No assignee Edit question
- Last query:
- Last reply:
Can you help with this problem?
Provide an answer of your own, or ask tgoeg for more information if necessary.