Release notes for MegaMatcher 12.x line products

These release notes are relevant for the following products:

• MegaMatcher 12.x Standard / Extended SDK
• VeriFinger 12.x Standard / Extended SDK
• VeriLook 12.x Standard / Extended SDK
• VeriEye 12.x Standard / Extended SDK
• VeriSpeak 12.x Standard / Extended SDK

See MegaMatcher 13 product line release notes for the newer versions.

See also the list of release notes for previous versions.

Product revision number: 20231030
Product revision hash: ebf47e2a4b922076b85d3143fdca228fd24c83f5

Fixes:

• Fixed a number of bugs and issues.

Updates and improvements:

• Expand standard record reading to handle more non-compliant data.
• Updated NdmMantraFinger biometric device support module.

Product revision number: 20230517
Product revision hash: 1b2938ddcbe61cffb568e4469cb2d244ae2432a6

Fixes:

• Fixed issue with building samples using wxWidgets v3.2.1 while maintaining backwards compatibility.
• Fixed issue with CSD101i finger scanner detection.
• Numerous bug fixes have been applied to the HFSecurity SDK, enhancing its stability.

Updates and improvements:

• A new configuration option for continuous device connection monitoring has been introduced to the NdmLumidigm finger module.
• Non-automatic (force) capture feature in NdmJenetric module is now supported.
• IntegratedBiometrics is now easier to use on x86 Linux.
• The camera on the IrisID TD100 iris scanner can now be used.
• Numerous enhancements have been made to the NdmMantraFinger module, making it more robust.
• New biometric capture devices are now supported:
  • BioID BioIris iris scanner.
  • Mantra MORPHS finger scanner.
  • NASPS NSP20i iris scanner.
  • NextBiometrics NB-2033, NB-65200 and NB-65210 finger scanners.
• Updated support modules and SDKs for biometric devices:
  • AratekTrustFinger SDK updated to v2.1.0.4.
  • Biometrika SDK updated to v2.4.2.
  • IntegratedBiometrics SDK from v3.9.1 to v3.9.2.
  • Jenetric SDK to v2.51.1.
  • NASPS SDK from v2.1.0.4 to v2.1.0.6.
  • NextBiometricsNB from v3.0.0_2633 to v3.0.0_2732.

Product revision number: 20221104
Product revision hash: dbbc26f6a1d09f5b8b47b5638b1aca7b6b8de4e5

Fixes:

• Fixed several minor bugs and issues.

Product revision number: 20221026
Product revision hash: 3e8f5ba1a532e6be1f787f2af9157df9b2d62f6e

Fixes:

• Fixed NSubject creation from FIRecord when images with segmentation are involved.
• Fixed issue with faces detection when face rectangle out of image bounds.
• Fixed several issues with wxWidgets faces sample.

Updates and improvements:

• Restored NDM module binary compatibility with the previous line. A scanner module from this product line can copied to the 11.2 line and will just work.

Product revision number: 20221024
Product revision hash: ff0a85d00503841faf088bcd66c263c3414d1321
The first release of MegaMatcher 12.4 SDK line.

Algorithms-related changes:

• Face liveness check:
  • The new highly accurate algorithm has been certified for a level 1 PAD by iBeta Quality Assurance. The certification includes smartphone and PC with camera scenarios, and ensures compliance with ISO 30107-3 Biometric Presentation Attack Detection Standards.
  • Specific parameters allow better algorithm adaptation for image and stream processing.
  • Quality check was integrated into liveness processing so both checks can be performed simultaneously.
• Fingerprint quality check:
  • New attributes which allow better control on biometric data quality and can tell if a finger is too wet, too dry, pressed too much or not enough, or if only tips of fingers are present.
• Iris quality check:
  • New attribute for detecting iris rotation (roll) angle in an image.

New features:

• Face to background ratio can now be specified for a token image via a parameter.

Product revision number: 20220928
Product revision hash: 5e6bffd386b7250d8bbb023582f63e61550abca9

New features:

• Added Heavy frame ICAO warning.
• Added single frame liveness in all ABIS and simple face/finger/iris samples.

Updates and improvements:

• FMRecord and ANTemplate: improved validation according to standard rules.
• JPEG file reading got better diagnostic messages for incorrect files.
• Android samples: removed audio permission where not needed and added Bluetooth permission where needed.
• Added fingerprint reader support:
  • Aratek A800
  • Tatvik TMF20
• Updated these included SDKs from biometric device manufacturers:
  • CredenceID SDK to v5.6.1.
  • Jenetric SDK v2.40 to v2.45.1.

Product revision number: 20220707
Product revision hash: 505ed7f747162606cd92ac9842d96ce56735d6ab

New features:

• Added samples for sending tasks to MegaMatcher Accelerator product using gRPC protocol.

Updates and improvements:

• Better recovery of ANTemplate files with incorrect header.
• Allow to read some broken JPEG files with invalid resolution.
• Improved NBiometricClient communication with MegaMatcher Accelerator to better handle timeout conditions.
• Added fingerprint reader support:
  • BioID BioSlap (new model)
  • Integrated Biometrics Columbo Mini
  • Suprema RealScan S60
• Expanded the support of fingerprint readers:
  • BioID scanners now supported on x86_64 platform
  • Futronic scanners now supported on macOS arm64e (Apple M1)
  • HFSecurity finger scanners now supported on Android
  • Integrated Biometric devices now have spoof finger detection on Android
• Updated these included SDKs from biometric device manufacturers:
  • Credence ID Android SDK to v5.5
  • DigitalPersona Android SDK to v3.2.5 which fixes bugs on newer Android versions
  • Integrated Biometrics SDK to v3.9.1 (Windows, Linux, Android)
  • Jenetric's SDK to v2.40
  • Suprema Realscan SDK (Windows and Linux) to v1.9.0

Product revision number: 20220317
Product revision hash: 67b9a34f65361de659f83b6c2fb24c7626b0ec15

New features:

• Added "passive with blink" face liveness check option to samples.

Fixes:

• Fixed issue with crash when performing voiceprint extraction on mobile platforms.
• Fixed several memory leaks in voice extraction.

Updates and improvements:

• Handle invalid WSQ NIST comments and fallback to 500 ppi resolution instead of failing with error.
• Better licensing support on Android 12.
• Aratek A900 fingerprint scanner is now supported.
• Updated Integrated Biometrics scanner support module with the newest SDK v3.8.0 (Windows and Linux). Spoof finger detection is now available.

Product revision number: 20220211
Product revision hash: 09c57792ec006759042f2a254dc5f0ccfe892336

Algorithms-related changes:

• Greatly improved faces passive liveness performance for mobile devices. Fixed some issues when devices are rotated.

Fixes:

• Fixed NFaceView attribute drawing on Android.
• Fixed trial/internet licensing crash when lots of file descriptors were opened on Linux.
• Fixed a number of minor bugs and issues in samples.

Updates and improvements:

• SecuGen Hamster Air (U-AIR) contactless fingerprint reader is now supported.
• Improved trial licensing support on modern Android devices.
• QoL improvements in samples.

Product revision number: 20220109
Product revision hash: 298625706268bd00794b137bae5a9f780779ce86
The first release of MegaMatcher 12.3 SDK line.

Algorithms-related changes:

• Improved extraction and attribute detection performance for all supported modalities.
• Fingerprints:
  • Integrated NFIQ 2.1 finger quality support.
• Faces:
  • Updated face detection algorithm, working better with cropped faces, masked faces, bad pose faces (profile).
  • Greatly improved face blink, dark glasses, expression, gender, glasses, mouth open, beard, mustache, hat attribute detection. Introduced heavy frame detection.
  • Better face liveness algorithm. Note: Passive mode was changed not to require blink, previous mode can be restored via a parameter.
  • Face passive liveness sensitivity can now be controlled by user via a parameter.
• Irises:
  • More accurate iris liveness detection.
• Voiceprints:
  • New voices algorithm. Note: templates are not compatible with previous versions of the SDK.

Fixes:

• Fixed a number of minor issues.

Updates and improvements:

• Added support for Apple M1 processors.

Product revision number: 20211018
Product revision hash: a106f2ee14cfe3d69961161efff37ba055c75b77

Algorithms-related changes:

• Improved background uniformity algorithm.
• Improved face background elimination algorithm accuracy and performance.

New features:

• Introduced possibility for existing iris templates to automatically determine position if it was not set already (Unknown).

Fixes:

• Fixed issue when creating token image with background removal and different width settings.
• Fixed crash in SimpleFingersSampleWX when loading 4 fingers slap image.
• Fixed several issues when multiple applications share the same dongle.
• Fixed issue with SDK component loading on some versions on macOS.
• Fixed issue with Android 11 Internet license usage.

Updates and improvements:

• Face background elimination now optionally supports alpha channel instead of setting constant background color.
• More user friendly messages for Trial/Internet license connectivity issues.
• Added BeepAfterScan custom property to NdmJenetrics.
• Updated Jenetric's SDK to version 2.29.1.937.
• Updated Integrated biometrics SDK to version 3.7.2.
• Updated Secugen Android device module.
• Updated Suprema BioMini SDK which supports more devices.

Product revision number: 20210804
Product revision hash: 87cab7ce292c62d945979996d61af99b3fb138b6

Algorithms:

• Added support for detecting liveness of iris images (paper prints, lenses with other iris pattern, etc.). The threshold can be adjusted via a parameter.
• Added support for automatic iris position detection.

Fixes:

• Restored missing NDF for faces OptimizedSpeed extraction scenario.

Updates and improvements:

• Added check for multiple faces in the image during face detection for scenarios when only one face should be present.

Product revision number: 20210709
Product revision hash: b9fa1fc33e4c9fc33b2397042abe809a6fa8fbe0
The first release of MegaMatcher 12.2 SDK line.

Algorithms:

• Faces:
  • Added the fastest extraction speed scenario for faces. It can be enabled using a parameter. This scenario is especially useful for devices with low compute power to improve extraction time. Obviously, this results in lower quality. Use this option only if default extraction speed was found unnaceptable.
  • New face liveness detection mode optimized for high resolution images. Can be enabled using using a parameter.
  • Face/not-face filtering for existing face templates. Latest face quality algorithm can be applied to existing templates generated even with previous versions of the SDK in order to accurately remove non-face images. This feature is enabled using a parameter. When enrolling face templates to engine, if a face does not pass quality threshold it means either there are no face or low quality face is present in template and ObjectNotFound NBiometricStatus is returned. It is possible to adjust the threshold.
  • New high quality face photo background elimination algorithm. For mobile devices there is a special version which works a lot faster but is less accurate. User may choose between high quality and fast version using a parameter.
  • More accurate face landmark detection.
• Irises:
  • Possibility to automatically determine iris position. It is beneficial to set position to left or right for irises so that matching is done only with compatible position resulting in better performance. This feature can be enabled using a parameter.
  • Improved iris extraction and matching algorithm. More accurate template is created when engine.TemplateSize=Large (default since this line). Iris templates are interoperable with the previous product lines.

New features:

• Added parallel face extraction tutorial for C language which demonstrates how to perform tasks in parallel for maximum performance.

Fixes:

• Fixed several issues in Activation Wizard.
• Fixed license generation bug on Android when using BouncyCastle as the security provider.

Updates and improvements:

• Better interoperability with older SDK versions when using dongle licensing.
• Better issue diagnostics message on Android if Internet permission not present.
• Fingerprint scanners:
  • Added DigitalPersona U.are.U support on ARM Linux.
  • Added support for Jenetric LIVETOUCH QUATTRO Compact scanner. Updated Jeneteric's SDK which includes advanced spoof detection and some bug fixes.
  • Added Mantra finger scanner module (MFS500 and MAPRO CX devices supported).

Product revision number: 20210415
Product revision hash: 4c39dd381b7ec473a4f7182b9f33967ae1be2857

Fixes:

• Fixed component signing issue on macOS.
• Fixed ZKTeco and AratekTrustFinger memory leaks.

Updates and improvements:

• Improved JPEG 2000 format support, including better handling for some invalid images.
• Reduced memory usage by not loading fingerprint scanner modules for not present devices.
• Better Android 11 support in samples and tutorials.

Product revision number: 20210326
Product revision hash: fdb5b0580f857cc83c69677b29448b9d1a5f03d6

Fixes:

• Fixed issue with face mask detection.
• Fixed minor licensing-related bug.

Product revision number: 20210323
Product revision hash: a5e41de93120d19f455a03aeea037f4f7c75ee7a

Algorithms:

• New fingerprint segmentation algorithm (first place in the NIST SlapSeg III evaluation, 2 inch category).
• Enhanced face detection algorithm to better filter out non-faces.
• Improved face quality assessment algorithm.
• Better single frame face liveness detection.
• Improved face mask detection.

New features:

• Predefined engine parameter sets optimized for particular use case are now supported.
• Face filtering by maximum pitch angle added.
• Color for face token image extrapolation can now be specified with the new Faces.TokenImageExtrapolationColor engine parameter.

Fixes:

• Fixed usage of Faces Min/Max IOD parameters and improved IOD estimation for rotated faces.
• Fixed NFaceView scaling with resolution issue.
• Fixed a number of other minor bugs and issues.

Updates and improvements:

• Updated IntegratedBiometrics SDK to 3.7.0 on Windows, Linux and Android.
• Updated MobotixThermal SDK to 1.1.0.5.
• Updated Jenetrics SDK to 2.22.0.

Product revision number: 20201111
Product revision hash: 414d47a2a5619dcac90ccbb6c6a9f59cc79c653e
The first release of MegaMatcher 12.1 SDK line.

Algorithms:

• Fingerprints:
  • Special fingerprint matching mode for the cases when some fingerprint records have incorrect resolution.
  • Fingerprint segmenter algorithm updated to improve handling of complex segmentation cases.
• Faces:
  • Face liveness detection from a single frame added.
  • Face mask detection accuracy improved.
  • Facial feature point detection algorithm accuracy improved.
  • Updated face liveness algorithm including passive and active modes and better blink detection.
• General:
  • Extraction algorithm performance greatly improved on Android, iOS and ARM Linux platforms.

New features:

• Added possibility to connect to MegaMatcher ABIS server using NBiometricClient. This allows easy migration of applications which were created to perform operations locally or communicate with MegaMatcher Accelerator. Abis Samples for all programming languages demonstrate this functionality in practice.
• Quality checks can now be disabled for a specific biometric modality by setting a parameter value. Previously the biometric template extraction algorithms by default performed quality checks and rejected poor quality biometric data. However, in some scenarios rescanning or rejecting data was not an option.
• WebP image format support added.

Fixes:

• Fixed a bug in fingerprint segmenter, which caused assigning unknown position if number of fingers is not as expected.
• Fixed the issue with facial feature points "jumping" between frames when processing video stream.
• Fixed issues in voices text dependent extraction.
• Fixed memory leak in voice extractor.
• Fixed licensing support for Android 11.
• Fixed Lumidigm 32 bit dll loading error.
• Fixed issue with Suprema Bio Mini SFR-410 device instantly capturing even with no finger on.
• Fixed the bug with duplicate SecuGen libraries loading.

Updates and improvements:

• Greatly improved face quality checking. Added new quality check failure status to indicate face covered with mask.
• WSQ 3.1 certification passed by Neurotechnology in 2021-01 on all supported platforms.
• Finger liveness parameters named in consistent pattern with face liveness instead of the previous misleading name.
• Integrated iMD GF601BM fingerprint module support.
• Updated Suprema RealScan SDK to version 1.8.7.6 with some small tweaks.
• Updated Jenetrics SDK to version 2.19.0.539 which fixes support for older CPUs.
• Updated HID Lumidigm device module to use the latest vendor SDK. New devices are now supported.
• Added SocketTimeout custom parameter to Futronic EthernetFam.

Product revision number: 20201111
Product revision hash: 414d47a2a5619dcac90ccbb6c6a9f59cc79c653e

New features:

• Added Android Bluetooth sample.
• Added spoof support and smear detection for Integrated Biometrics scanners on Microsoft Windows.

Fixes:

• Fixed a bug related to face extraction from stream operation which could never finish with bad quality input.
• Fixed an issue with outside image bounds face detection.
• Fixed an issue with Nfiq2.framework signing for macOS.
• Fixed issues with license deactivation on older Android devices.
• Fixed a number of bugs and issues in the iOS biometric sample.
• Fixed a bug related to Upek (and other DigitalPersona) device capture on Windows and Linux.
• Fixed an issue with Integrated Biometrics automatic scan after manual.
• Fixed an issue supported device list in algorithm demo.

Updates and improvements:

• Improved fingerprint segmentation.
• Better "main" face selection when multiple faces are present in image. Criteria include face quality, face size, face centerness.
• Optimized NImage creation from bitmap on Android.
• Added support for iMD SF202 and SF302GM fingerprint sensors.
• Added support for NextBiometrics NB-65200 fingerprint module.
• Added support for Thales Cogent CSD101i (aka Green Bit CSD101i) on Windows.
• Added support for Suprema RealScan FC palmprint scanner.
• IrisId major stability rework.
• Added Futronic scanners support for Android 10/11.
• MobotixThermal camera can be added through the configuration file.
• Faster capturing on Aratek BM series devices.

Product revision number: 20200928
Product revision hash: df75555f0699e25bc84259d1db9ada9c3d914018

The first release of MegaMatcher 12.0 SDK line.

Algorithms:

• Fingerprints:
  • NFIQ 2.0 support – NIST Finger Image Quality (NFIQ) is a standardized fingerprint image quality metric used to determine if samples are of acceptable quality to be used for recognition.
  • Finger presentation attack detection for rejecting templates extracting after scanning spoof fingers.
  • Finger slap image quality assessment to detect incorrectly scanned images, such as only finger tips present or wrong hand or thumbs in a 4-4-2 scenario.
• Faces:
  • New generation face detection algorithm which allows wider face position tolerance as well as general enhancement to recognize faces with face masks.
  • New facial feature points detection algorithm is more accurate, especially in more difficult cases like non-frontal images, exaggerated facial expression or harsh lighting conditions.
  • New face recognition models introduced to provide higher matching accuracy.
• Voiceprints: improved text-independent recognition.

New features:

• Licensing high-availability – multiple license servers can be used in such way that when at least one license server is functioning properly, application using licenses will continue to work. Also, the solution takes care of lot of details, such as automatically retrying to reobtain licenses which became temporary unavailable

Updates and improvements:

• Faster image reading. WSQ reading speed increased 3 times on PC and 2 times on ARM. TIFF image reading was optimized. Some other common cases were also optimized.
• Java version was upgraded to Java 8 (previously was 6). JNA version upgraded to 5.5.
• NVideoWriter was dropped (deprecated several years ago).