KYC-Documents-Verif-SDK - wrong class_name

[Zux BG]

Hi,

I've tested 5 ID cards of Republic of Serbia, and for all of them online test tool (https://www.doubango.org/webapps/kyc-documents-verif/#) reports as

class_name :  "Serbia - Id Card (2009-2014) #1"

while these ID cards are valid and of newer type. All sample IDs return this fragment.

• VERIF : [
  • {
    • description :  "Serbia - Id Card (2009-2014) #1",
    • fields : [
      • {
        • VIZ_INVALID_FIELDS : [
          • "Sex"
          ]
        },
      • {
        • VALIDITY : [
          • "VIZ_INVALID"
          ]
        }
      ]
    }
  ],

and

• {
  • box : [ ... ],
  • name :  "Sex",
  • ocr_lang :  "eng",
  • ocr_pass :  "VIZ",
  • ocr_score :  0.93,
  • ocr_text :  "Ж",
  • ocrable :  true,
  • rotation :  0,
  • valid :  false
  },

So, the field "sex" is OCR-ed OK, and seen here in Cyrillic alphabet.

I see two errors:

1. Wrong class_name :  "Serbia - Id Card (2009-2014) #1" - it should be newer ID type class.

2. "Sex" validation if false negative.

[Mamadou DIOP]

Hi,

We support 2 Serbian ID cards: 

- "Serbia - Id Card (2008)" -> https://www.consilium.europa.eu/prado/fr/SRB-BO-01001/image-354092.html

- "Serbia - Id Card (2009-2014) # 1" -> https://th.bing.com/th/id/R.8393781734fff926f684e06e304d4f60?rik=bLqmCMJvrBOykQ&riu=http%3a%2f%2fwww.e-ducan.com%2fassets%2fimg%2fusers%2fIDs%2fDSC_0787.JPG&ehk=Pd3x0Q0nBorTEVCEZbuBrMMIgcW9qm79t4bIgYOWd3c%3d&risl=&pid=ImgRaw&r=0

Full list at https://github.com/DoubangoTelecom/KYC-Documents-Verif-SDK/blob/main/Graphs.pdf

I'm not aware of any new one.

The "Sex" being returned as invalid is an annotation issue. We reported it with regular expression "\p{Lu}/\p{Lu}" instead of ""\p{Lu}" which means we expect "Ж/F" or "M/M" instead of "Ж" or "M". I have fixed it, will be part of next release.

--
You received this message because you are subscribed to the Google Groups "doubango-ai" group.
To unsubscribe from this group and stop receiving emails from it, send an email to doubango-ai...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/doubango-ai/4d027a3b-4d13-4edd-8246-d919b98e8bfcn%40googlegroups.com.

[Zux BG]

Thanks for quick answer. Yes, this could be the case for older ID card types, to require sex abbreviation on both Latin and Cyrillic ( "Ж/F" for female, for example). Now, it is only in one alphabet, default in Cyrillic, optionally in Latin. Here is sample specimen, but not official (there is no official one):   Serbian identity card - Wikipedia     
So, after 2014, VIZ is changed in Sex field (only one letter).

2014-2015 new chip, not important for OCR

2021- digital signature changes,  not important for OCR

I am happy to test new releases.