Displaying present location in the site.

World's best face recognition technology achieves a safe and secure society

Face recognition technology is used in a wide range of areas spanning an array of industries, including national security infrastructure applications such as immigration control and national IDs, security enhancement applications such as entrance and exit control and mobile phone and computer login security, as well as applications to enhance the sophistication of corporate services, and equipment embedding applications.
NEC has continuously worked to refine its face recognition technology since beginning research and development in 1989, and has succeeded in achieving the world's highest performance evaluation for its still image and video face recognition technology, in benchmark testing conducted by the U.S. National Institute of Standards and Technology (NIST), which is recognized as a worldwide authority.

By summarizing information from features and essays that were published in past editions of NEC Technical Journal, this article provides an introduction to NEC's face recognition technology, which is proudly regarded as world-class, and looks back on the history of its research and development.

R&D began approximately 30 years ago, when practical application was considered difficult

NEC began research and development of face recognition technology in 1989. The research incorporated pattern recognition technology that was established through character recognition research over 50 years ago. At the time, countries around the world were conducting research in biometric authentication. However, even with the development of fingerprint authentication technology, it was still considered difficult to achieve face recognition technology that could be used in practical applications. NEC pursued efforts in basic research for a long time. Then, in 2009, in response to demand for advanced security measures following the 9/11 terror attacks in the United States, NEC began making serious efforts to apply the technology to products.

Cover of NEC Technical Journal, Vol.56 No.10
Cover of NEC
Technical Journal,
Vol.56 No.10

NEC unveiled its NeoFace system, which incorporates a unique combination of face detection technology that detects and extracts face regions from images, along with face matching technology that can determine the identity of the faces in the extracted images. NeoFace attracted a great deal of attention as a groundbreaking product with high-speed, high-accuracy face detection performance, as well as reduced sensitivity to environmental conditions such as variations in face orientation or changes in the lighting, which were considered weak points in conventional face recognition technology.

Three-time consecutive achievement of world's highest performance evaluation in benchmark tests by worldwide authority

With more and more attention being paid to public safety around the world, NEC developed technology that further improves face recognition accuracy by analyzing captured images and previously registered candidate images, and matching the characteristic facial features. This technology determines the face regions in facial images by performing face detection processing, and then calculates the positions of facial feature points such as the eyes, nose, and mouth corners. After the face region position and size are normalized using the obtained feature point positions, the matching process is performed.
In the still image face recognition category of the NIST technology evaluation program, this NEC face recognition technology demonstrated much better performance than the engines developed by the other participating organizations, thereby attracting attention from around the world. NIST has conducted the technology evaluation program since 1993, and it has become a benchmark test of worldwide authority that is employed even as a bid tender condition for immigration control systems.

NEC's Face Recognition Technology and Its Applications

With the recent rise in crime and terrorism the need for biometric personal authentication technology increases year by year. Face recognition technology has advantages not found in other biometric personal authentication technologies, e.g. a person does not have to take any special actions for authentication, and we human can identify the person visually with his or her face image. NEC’s Information and Media Processing Labs participated in the Multiple Biometric Evaluation program, a biometrics technology evaluation program of the National Institute of Standards and Technology (NIST), USA, that was held from January to May 2010. NEC here achieved the highest recognition performance among the participating organizations, with the highest identification rate of 95% of 1.8 million persons and the lowest verification rate of 0.3%. This paper describes the face recognition technology developed by NEC and introduces the evaluation results by NIST as well as discussing actual applications of NEC’s face recognition engine.

IMAOKA Hitoshi, HAYASAKA Akihiro, MORISHITA Yusuke, SATO Atsushi, HIROAKI Toshihiko
(NEC Technical Journal Vol.3 No.3 October, 2010)

NEC continued to develop its face recognition technology and continuously participated in the NIST benchmark tests, which serve as an objective indicator of performance evaluation. As a result, NEC's face recognition technology achieved the top ranking in all three consecutive years of testing (2009, 2010 and 2014). NEC's face recognition technology has obtained top ratings in many other areas in addition to recognition performance and speed. These include the ability to handle facial variations due to aging, which is required for recognizing people from their passport photos, the ability to handle different racial backgrounds, which is required in global business operations, and the ability to recognize faces oriented at various angles, such as tilted to the side or facing upward, which is required for surveillance of entrances and exits at key facilities.
A report published by NIST in 2013 assessed that NEC's algorithms have consistently maintained the highest accuracy since 2010, that NEC's error rate is the lowest of all the databases, and that NEC's search error rate is less than half that of the next-closest competitor.

The World's Best Face Recognition System to Achieve Safety and Security in Our Society

Computerization and the consequent diffusion of network services have been increasing the need for biometrics year on year. Face recognition is a technology that imposes a low work load on users because it is based on natural authentication movements, such as viewing a camera. NEC Information and Media Processing Laboratories first joined the Facial Recognition Evaluation Program of the U.S. National Institute of Standards and Technology (NIST) in 2009.Top ranking in recognition accuracy and search speed tests was awarded in the three consecutive years; 2009, 2010 and 2013. This paper describes the NEC face recognition technologies that achieved the top ranking performances, and discusses the results achieved in the NIST evaluation program of 2013.

IMAOKA Hitoshi
(NEC Technical Journal Vol.9 No.12 January, 2014)

Achievement of world's top ranking in video face recognition technology that enables real-time surveillance

With the widespread adoption of face recognition technology, there is a growing need for video face recognition technology that enables real-time surveillance. Still image face recognition is an active authentication procedure that requires subjects to voluntarily perform authentication by standing in front of a camera. In contrast, video face recognition can be performed even without the subject's awareness of the camera, so the conditions of the environment (camera installation position, image quality, lighting conditions, subject size, etc.) and subject movement (walking speed, face orientation, eye direction, etc.) can have a significant effect on the performance. As such, the technology needs to be far more advanced than that of still image face recognition. In response to this issue, NEC incorporated deep learning into its technology for identifying individuals by their facial features. This provided the capabilities to handle variations in face orientation and low-resolution facial images, thereby achieving a high degree of accuracy in video face recognition.
In video face recognition technology benchmark tests conducted by NIST in 2017, NEC's video face recognition technology earned first place in the performance evaluation with a matching accuracy of 99.2%, which was dramatically higher than the competitors. With this achievement, NEC's face recognition technology PDFtook first place for the fourth consecutive time, following the previous achievements in still image face recognition technology.

Video Face Recognition System Enabling Real-time Surveillance

The recent sharp increase in terrorist attacks has made the security of public facilities such as airports, restaurants and hotel lobbies a global issue. The face image in a passport is an only data to be used for personal authentication worldwide, regardless of nationality. In addition, the face image recognition offers a unique advantage compared to other personal authentication technologies because it enables to authenticate people from a distant location such as personal identification through surveillance cameras. NEC joined the Face Recognition Vendor Tests of the U.S. National Institute of Standards and Technology (NIST) in 2009 and gained the top ranking three times consecutively. This paper introduces the NEC video face recognition technology for use in video surveillance.

IMAOKA Hitoshi
(NEC Technical Journal Vol.11 No.1 December, 2016)

With rising concerns about security around the world, face recognition technology is becoming a critical component of the security infrastructure that is essential to public safety.
NEC's face recognition technology related products have already been adopted in more than 100 systems in 40 countries across the globe. These include security measures for some of the world's major organizations, such as JFK Airport in New York, Australian governmental institutions, and major airports in Brazil. NEC's face recognition technology has also demonstrated its effectiveness in the field of entertainment. For example, it has been used at 50 music concerts and sporting events in Japan for audiences numbering more than 300,000.

NEC is currently focused on the social solutions business, including efforts to strengthen the safety business as a pillar of the global growth strategy. Moving forward, NEC will remain focused on developing solutions that take advantage of face recognition technology, which is one of the cutting-edge AI technologies in the NEC the WISE portfolio, with the aim of contributing to safety and security in a wide range of fields, including social infrastructure and private sector facilities.

Related link

Check out NEC's latest efforts to support Face Recognition.

(Published September 28, 2018)