This is the top of the page.
Displaying present location in the site.
  1. Home
  2. Products & Solutions
  3. Public Safety
  4. By Technology
  5. NIST Proven Accuracy
Main content starts here.

By Technologies

NIST-Proven Accuracy

NEC’s unsurpassed face and fingerprint matching algorithm provides high accuracy and selectivity regardless of the database size and image quality. Through a comprehensive range of tests, from small one-to-one verification all the way up to large-scale, high-volume identification matching, conducted by the National Institute of Standards and Technology (NIST), the results have validated NEC AFIS as the best in the industry.

Working with NEC means having access to decades of biometrics expertise and proven biometrics technologies that ensure performance, accuracy and reliability.

Face Recognition

Fingerprint Recognition

Note: NIST test results do not constitute endorsement of any particular system by the government. For more information, visit www.nist.gov.

Face In Video Evaluation 2017[New]

The Face in Video Evaluation (FIVE) was conducted to assess the capability of face recognition algorithms to correctly identify or ignore persons appearing in video sequences. Using video images from standard cameras for face recognition requires highly-advanced techniques when compared to still images. This is because images are greatly influenced by environmental conditions, such as camera location, image quality, lighting and subject size, in addition to the behavior of a subject, including walking speed, face direction and sight line. To achieve reliable face recognition of a video image, NEC developed feature point extraction technology that enables enhanced face recognition to a level where an individual can be identified with high precision from within a group, even if their face is partially hidden, or the image is taken from different angles. NEC's face recognition technology also uses deep learning technologies for face matching to increase accuracy to a level where an individual can be identified by a low resolution face image captured by a distant camera.

NEC ranked No.1 in accuracy in both ideal and difficult environmental conditions for face recognition.

FpVTE2012 – Fingerprint Vendor Technology Evaluation 2012

In FpVTE2012, which, unlike FpVTE2003, was intended to evaluate on operational datasets of an unprecedentedly large scale to reflect real-world applications, NEC AFIS achieved the highest one-to-many identification accuracy, thereby proving its feasibility and superiority.

NIST MBE Multiple Biometric Evaluation 2010

NEC face recognition technologies were ranked No. 1 in the MBE Still-Face Track in 2010 carried out by the National Institute of Standards and Technology (NIST), commissioned by the Department of Homeland Security. MBE's Still-Face Track benchmark test employed accuracy evaluation tests with a dataset of over one million face images, which were collected from actual criminal databases and face images used for visa applications. NEC engines are shown as Vxx in the following figures.

Highest identification rate

Highest 95% identification rate from 1.8 million visa applicants.

Lowest False Reject Rate 0.3% in processing the civil database at False Accept Rate of 0.1%.

Highest 92% identification rate from a 1.6 million person criminals.

Lowest False Reject Rate 4% in processing the criminal database at False Accept Rate of 0.1%.

Lowest verification error rate

Widest acceptance performance against yaw-angle variation.

Most robust against time-elapsed between photographs.

Most robust against short inner-eye distance.

Most robust against high compression rate (JPEG, inner-eye distance=96 pixels).

Most robust against high compression rate (JPEG2000, inner-eye distance=96 pixels).

Highest query processing speed by using 16 cores parallelly.

(1.6 million images can be processed in approx. 0.4 second)

For more information, please refer to the reports below.

NIST MBGC Multiple Biometric Grand Challenge 2009

NEC face recognition technologies were ranked No. 1 in the MBGC "Still-Face" Challenge Problem in 2009, carried out by the National Institute of Standards and Technology (NIST), commissioned by the Department of Homeland Security.

The facial images evaluated in the Still-Face Challenge Problem were taken by high resolution digital cameras under a variety of challenging conditions, including compressed images used for IC passports, and images taken under poor indoor lighting or direct sunlight. These situations were designed based on anticipated real-world scenarios.

At this test, NEC's superior test results have demonstrated highly accurate performance. Its False Reject Rate of 2.1% and False Accept Rate of 0.1% for outperformed technologies competing in the same category.

Highest verification rate

High tolerance to variations in illumination and compression

Lowest false reject/accept rate

0.3% False Reject Rate in processing visa application database
0.1% False Accept Rate in processing visa application database

For more information, please refer to the report below.

ELFT07 Phase2 - An Evaluation of Fingerprint Technologies 2007

NEC's latent fingerprint matching technology was ranked number one in the world on the Evaluation of Latent Fingerprints Technologies (ELFT) test carried out by the National Institute of Standards and Technology (NIST), commissioned by the U.S Department of Homeland Security.

Among the world's leading fingerprint matching system vendors participating in this evaluation, NEC far surpassed even the second ranking vendor and reassured the superior position of NEC's fingerprint technology. NEC's high accuracy rate is achieved by the integration of improved matching technologies, such as a low-confidence minutiae adaptive matching algorithm, a zone matching algorithm, and recently developed image processing technologies specialized for latent fingerprints, such as a latent background noise removal algorithm and a low-quality fingerprint ridge recognition algorithm.

The ELFT07 was a test of accuracy for searching latent fingerprints when using automatic feature extraction and matching, conducted in the United States in 2007 under the control of one of the US’s most respected government authorities, the NIST.

For more information, please refer to

MINEX04 - Minutiae Interoperability Exchange Test 2004

The growing use of biometrics for security applications in immigration and border control has focused attention on the importance of interoperability among multiple matching systems. The MINEX, conducted by NIST and sponsored by the Justice Management Division of the U.S. Department of Justice and the Department of Homeland Security US-VISIT program, determines the feasibility of using minutiae data for the exchange of fingerprint information between dissimilar fingerprint matching systems.

In MINEX04, NEC AFIS was evaluated as one of the most accurate algorithms in the minutiae data exchange interoperability test. Where vendors own proprietary minutiae data was used, NEC AFIS earned the top ranking (i.e., proved overall superiority).

For more information, please refer to the report below.

SlapSeg04 - Slap Fingerprint Segmentation Evaluation 2004

In SlapSeg04, NEC AFIS achieved the top score in segmentation accuracy. Additionally, NEC AFIS proved most accurate in hand and finger position identification.

Slap fingerprints (or slaps) are plain four-finger images 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. The high accuracy of slap segmentation contributes to the total system cost performance.

The SlapSeg04 was conducted by the NIST in collaboration with 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). This evaluation determined the accuracy of the slap fingerprint segmentation algorithm (i.e., one component of AFIS) used to segment slap fingerprint images into individual fingerprint images.

For more information, please refer to the report below.

FpVTE2003 - Fingerprint Vendor Technology Evaluation 2003

In FpVTE2003, NEC AFIS earned the top ranking for its unparalleled matching accuracy in all accuracy tests using real data sets. To be particularly noted is that, even when the data had poor quality, NEC AFIS showed remarkable performance standing out from all competitor AFIS systems by a substantial margin.

The FpVTE2003 was an international benchmark test of fingerprint matching, identification, and verification systems, conducted in the United States in 2003 under the control of one of the US’s most respected government authorities, the NIST.

For more information, please refer to te report below.

Contact

Top of this page