Neurotechnology released new versions of the company's product line of SDKs for biometric fingerprint, face, iris and voice identification and object recognition.

The updated product line includes new embedded versions of VeriFinger, VeriLook and MegaMatcher as well as the introduction of VeriEye Embedded for iris, VeriSpeak Embedded for voice and SentiSight Embedded object recognition technologies for use in smartphones and other mobile devices.

"Our new product line includes updates to all of our main SDKs, allowing customers to develop faster and more reliable applications," said Algimantas Malickas, CEO of Neurotechnology. "The new versions of our embedded SDKs can be used on Android devices for development of a full-range of single or multi-modal biometrics applications and object recognition systems, from simple smartphone applications to large governmental projects."

Updates to Neurotechnology's biometric and object recognition algorithms provide a number of enhancements over previous versions.

VeriFinger 6.5 SDK for fingerprint identification includes significant enhancements over the previous version in terms of recognition reliability, with a 20% decrease in the false rejection rate (FRR) when set at the same false acceptance rate (FAR). The new VeriFinger Embedded 1.1 SDK uses the new VeriFinger 6.5 algorithm.

VeriLook 5.2 SDK for facial identification also includes improvements in recognition reliability with particular enhancements for recognition of faces that are rotated out-of-plane. Both VeriLook Embedded 1.1 SDK and the new VeriLook Surveillance 2.1 SDK use the new VeriLook 5.2 algorithm.

VeriEye 2.5 SDK for iris identification includes a faster algorithm and a new template extraction option. With the new VeriEye Embedded 1.0 SDK, based on the VeriEye 2.5 algorithm, iris recognition is now available for use in embedded applications.

VeriSpeak 1.1 SDK includes the addition of new pause detection technology and the ability to recognize both unique pass-phrases for each user as well as fixed phrases – phrases that are the same for all users – by recognizing the user's unique voice. With the introduction of VeriSpeak Embedded 1.0 SDK that utilizes the VeriSpeak 1.1 algorithm, voiceprint identification is now possible for use in smartphones and other mobile devices.

MegaMatcher 4.3 SDK for large-scale biometric solutions includes all improved algorithms, making it possible to develop multi-biometric solutions using any combination of fingerprint, face, iris and voice identification technologies for added accuracy and reliability. MegaMatcher Embedded 1.1 SDK includes all new embedded versions.

SentiSight 3.1 SDK for object recognition and computer vision includes a new algorithm that is twice as fast as the previous version and now provides the ability to track multiple objects at the same time. The new SentiSight 3.1 algorithm is now available for the first time in an embedded version, SentiSight Embedded 1.0 SDK, for the development of object recognition, robotics, augmented reality and mobile law enforcement applications such as tattoo recognition.

All embedded SDKs are designed for the Java development environment and support the Android operating system.