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
This commit was created on GitHub.com and signed with GitHub’s verified signature.
The key has expired.
New features
We translated complete SDK to the following additional languages: Malay, Dutch, Hungarian, Slovenian, Indonesian, Arabic(UAE), Romanian, Chinese traditional, Chinese simplified, Thai, Hebrew, Vietnamese, Filipino.
Improvements to existing features:
We have improved parsing of MRZ formats deviating from the ISO/IEC 7501 standard:
Document discriminator was in place of the document number on driver licenses and IDs from:
New York
Michigan
Canada
Different check digit calculation for Mexico (Consular) ID
Recognition of the unofficial XCT country code for Northern Cyprus ID
Recognition of different country codes and check digit calculation on China Mainland Travel Permit for Hong Kong and Macao Residents
We added anonymization support for:
MRZ on the Mainland Travel Permit for Hong Kong and Macao Residents
We have made some changes to the BlinkID(Combined)Recognizer
You can now see ProcessingStatus in the results to inspect potential processing errors, such as when barcode detection fails, a mandatory field is missing, etc.
You can now also see a more detailed ImageAnalysisResult showing you when:
Face image is detected
MRZ is detected
Barcode is detected
We added a RecognitionModeFilter settings group. You can toggle flags on this object to control the recognition mode of the recognizer:
enableMrzId lets you scan MRZ on all identity documents except visas and passports.
enableMrzVisa lets you scan MRZ on visa documents.
enableMrzPassport lets you scan MRZ on passports.
enablePhotoId lets you scan photo IDs. Use it to enable or disable document and face image extraction on unsupported documents.
enableFullRecognition lets you scan all data from our supported documents.
Your license key still controls which of the above recognition modes are allowed.
We have added a RecognitionMode result member describing which recognition mode was used to produce the results.
We are now retrieving sex and nationality fields from the MRZ in cases where those two fields cannot be found in the document’s VIZ. Previously, we only used to do this for dates, name fields and document numbers.
We are now preserving the original string (raw data) of the dates we couldn’t parse.
We have improved the thresholds for card detection feedback messages ("move closer" and "move farther"). This will improve the UX when scanning in landscape mode as the document can now be closer to the camera.
Minor API Changes
We have made some changes to the BlinkIDRecognizer and BlinkIDCombinedRecognizer:
We renamed DocumentImageMoireStatus to ImageAnalysisDetectionStatus
We grouped the conditions member from the results with the DriverLicenseDetailedInfo structure
We renamed RecogitionMode to RecognitionDebugMode in RecognizerBundle.
Bug fixes:
We fixed an issue in default UI that was causing reticle to reappear while card flip animation was showing.
We fixed an issue that was causing some methods marked with @NonNull to return null.