You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is it possible to do OSD but specify the script to improve the chances of correct orientation?
I'm scanning identity documents, which will always be Latin characters. But it seems OSD is thrown off by the presence of a MRZ (machine-readable zone), which you may have seen in passports and other IDs, which contains <<<<<<<<< fillers for spaces.
The side of the identity document that contains no MRZ seems to be reliably detected.
The only solution I found so far is to just brute-force it until I found a string of <<<<< (which unfortunately, very often gets confused for L most of the time).
The text was updated successfully, but these errors were encountered:
Is it possible to do OSD but specify the script to improve the chances of correct orientation?
I'm scanning identity documents, which will always be Latin characters. But it seems OSD is thrown off by the presence of a MRZ (machine-readable zone), which you may have seen in passports and other IDs, which contains
<<<<<<<<<
fillers for spaces.The side of the identity document that contains no MRZ seems to be reliably detected.
The only solution I found so far is to just brute-force it until I found a string of
<<<<<
(which unfortunately, very often gets confused for L most of the time).The text was updated successfully, but these errors were encountered: