NEC (UK) Ltd has today announced that NECs fingerprint technology earned top ranking in most categories of the Slap Fingerprint Segmentation Evaluation 2004 (SlapSeg04), sponsored by the Justice Management Division of the U.S. Department of Justice.
Slap fingerprints (slaps) are taken by pressing the four fingers of one hand onto a scanner or fingerprint card simultaneously. Slap segmentation is the process by which a single image containing four fingerprint images is divided into four images of the individual fingers. Slap fingerprints are noted for the speed at which they can be documented and processed, and their use is being considered for background checks in a variety of U.S. Government fingerprint systems.
The purpose of SlapSeg04 was to evaluate the accuracy of the slap fingerprint segmentation algorithm, one component of automated fingerprint identification systems (AFIS). The study was conducted by the National Institute of Standards and Technology (NIST). Additional partners of SlapSeg04 include the U.S. Visitor and Immigrant Status Indicator Technology (VISIT) Program Office, the U.S. Department of Homeland Security (DHS), and the Federal Bureau of Investigation (FBI).
NEC is pleased to once again have the accuracy of its fingerprint segmentation algorithm validated in a rigorous, controlled environment, as it confirms the success that our law enforcement customers in North American have been enjoying for over 20 years, Barry Fisher, associate vice president, AFIS Division, NEC Solutions America. We are proud that NECs technology earned top ranking among the competition.
The study was performed using almost 30,000 slap prints submitted from seven organisations.
NECs slap segmentation algorithm achieved a 96.8 percent correct segmentation rate when three or more highly-matchable fingerprints (and correctly identified finger positions) were measured, an accuracy rate 1.1 percent higher than that of the second ranked vendor. NEC also achieved a correct segmentation rate of greater than 99 percent in circumstances in which two or more marginal and highly-matchable fingerprints (and correctly-identified finger positions) were measured.
Additionally, NECs slap segmentation algorithm proved most accurate in hand and finger position identification rate for slaps with an unknown hand type. NEC correctly identified hand and finger position 99.8 percent of the time, 1.2 percent higher than the second ranked vendor. This translates to an error rate (.2 percent) one-seventh of the next-ranked vendors rate of 1.4 percent.
This accuracy rate can help detect uncooperative subjects who attempt to avoid identification by, for example, scanning interchanged hands. According to the study, only NECs slap segmentation algorithm can detect such illegal attempts with acceptable accuracy. This is the second time in less than one year that NEC has earned the top spot in a U.S. Government agency-sponsored fingerprint technology evaluation. In July 2004, NEC earned top ranking in all categories of the Fingerprint Vendor Technology Evaluation (FpVTE) 2003, sponsored by the Justice Management Division of the U.S. Department of Justice. The FpVTE evaluated the accuracy of fingerprint matching, identification, and verification systems, and was also conducted by the National Institute of Standards and Technology (NIST).
NEC's small-scale fingerprint identification system achieved an accuracy rate of 99.9 percent, confirming the extreme effectiveness of its fingerprint matching technology. Accuracy rates for NEC's medium-scale and large-scale systems were 99.4 and 99.3 percent, respectively.
NEC began research on fingerprint identification technologies in the 1970's. In 1982, NEC commercialised the relational matching formula for the first time in the world. NEC's automated fingerprint identification systems are currently used at law enforcement agencies in California and Indiana, and immigration offices in Singapore, and are presently being trialed throughout parts of the UK and Europe for National ID card and e-border projects.
Many public and private sector organisations rely on NEC's fingerprint matching technology to resolve their identity management problems. By combining fingerprint identification with other technologies such as SmartCards and digital certificates, NEC's advanced digital identity solutions provide organisations with key benefits such as network access, remote network access, physical access, signing and encryption services and secure single-sign-on capabilities.
The SlapSeg04 was conducted during late 2004 and early 2005. Ten vendors and organisations participated in the SlapSeg04 submitting a total of 13 algorithms for evaluation. All participants sent their slap segmentation software to NIST. The evaluation was done by SlapSeg04 staff at NIST facilities using more than 29,000 slap fingerprints.