MegaMatcher Palm Print biometric technology and components are intended for biometric systems developers and integrators who need to add palm prints to the list of supported biometric modalities. The palm print template extraction and matching components are available for MegaMatcher 4.2 SDK customers as part of the SDK distribution.

Palm Print Matcher component

The Palm Print Matcher component performs palm print template matching in 1-to-1 (verification) and 1-to-many (identification) modes.

One license for the Palm Print Matcher component is included in MegaMatcher 4.2 Standard SDK. Two licenses for the Palm Print Matcher component is included in MegaMatcher 4.2 Extended SDK. More licenses for this component can be purchased any time by MegaMatcher 4.2 SDK customers.

Palm Print Client component

The Palm Print Client component creates palm print templates from palm images. Also, it allows to integrate support for palm print template and image format standards and additional image formats with new or existing biometric systems based on MegaMatcher SDK.

These biometric standards are supported:

• ANSI/NIST-ITL 1-2000 (Data Format for the Interchange of Fingerprint, Facial, & SMT Information)
• ANSI/NIST-ITL 1-2007 (Data Format for the Interchange of Fingerprint, Facial, & Other Biometric Information)
• ANSI/NIST-ITL 1a-2009 (Data Format for the Interchange of Fingerprint, Facial, & Other Biometric Information)

The Palm Print Client component allows conversion between Neurotechnology proprietary palm print templates and ANSI/NIST-ITL templates.

The Palm Print Client component also includes:

• WSQ (Wavelet Scalar Quantization) image format support module. The WSQ format allows to compress a palm print image up to 10-15 times. WSQ compression process is "lossy", meaning that the reconstructed image is not equal to the original (some information is lost). However, the WSQ algorithm was specially designed to minimize the loss of palm print or fingerprint information therefore the reconstructed image is as close as possible to the original.
• JPEG 2000 image format support module .

The Palm Print Client component can be used from C/C++ and C# applications on all supported platforms. .NET wrappers of Windows libraries are provided for .NET developers.

One license for the Palm Print Client component is included in MegaMatcher 4.2 Standard SDK and MegaMatcher 4.2 Extended SDK. More licenses for this component can be purchased any time by MegaMatcher 4.2 SDK customers

Technical Specifications

All specifications are given for Intel Core 2 Q9400 processor running at 2.66 GHz.

Full palm print;
fingerprints marked in red for reference

Palm print template extraction and matching require much more time than fingerprints, as palm images are much larger compared to fingerprint images, but have similar features density.

An image of fingerprint, which was scanned with AFIS-class scanner at 500 dpi resolution, is usually at least 500 x 500 pixels (0.25 Megapixels). Full palm image, scanned at the same resolution, is 160 times bigger (40 Megapixels). After excluding white space, palm image is still about 50 times bigger than fingerprint image. Also, full palm print templates may contain about 2,000 minutiae compared to about 50 for fingerprint templates.

MegaMatcher palm print template matching algorithm can use more than one processor core on multi-core processors allowing to increase template matching speed. The template extraction algorithm can be also run on more than one processor core allowing to process several templates in parallel.

MegaMatcher palm print identification algorithm has this performance when processing full palm prints:

• Template extraction time: 6.8 seconds;
• Template matching time:
  • 3 templates per second (using 1 processor core);
  • 12 templates per second (using 4 processor cores);
• Average template size: 69 kilobytes.

Reliability Tests

Click to zoom

The MegaMatcher palm print template matching algorithm reliability tests were performed using internal palm print images database. The database contained 1,993 images of right hand full palms and 1,996 images of left hand full palms. The database represented 1,000 unique persons.

Receiver operation characteristics (ROC) curves are usually used to demonstrate the recognition quality of an algorithm. ROC curves show the dependence of false rejection rate (FRR) on the false acceptance rate (FAR). The chart with ROC curves for the MegaMatcher palm print template matching algorithm are available on the right.

Go to MegaMatcher contents