Aiming to “Connected World” from “Palm Vein Authentication” to “Multimodal Biometric Authentication”
Biometrics are drawing more attention as a highly convenient and secure method of authentication, and we often see this technology used in daily life. Fujitsu Laboratories takes this technology one step further by conducting research on ” Multimodal Biometric Authentication” which combines palm vein authentication and face authentication. What are the benefits that the practical application of this technology would provide for individuals and society at large? To this question, four researchers told the story of their efforts in which they played different roles.
Limitations of Biometric Identification
Biometrics is now widely used in everyday life, for example to unlock smartphones and laptops and so on. Among many different types of authentication, such as fingerprint authentication, facial authentication, and palm vein authentication, Fujitsu Laboratories has been focusing on developing palm vein authentication technology.
This technology was commercialized in 2003 as a technology to read and identify each person’s palm veins pattern by using near infrared light. Since then, its contactless and stable accuracy, etc. have been highly recognized, and many banks in Japan and overseas started to use this technology to authenticate ATM users. Now it is the most widely used authentication method in the world with 94 million users in 60 countries.
Actually, using one’s biometric information only, such as fingerprints and veins, it will be possible to identify the person, because biometric information is unique to that person. However, biometrics are generally used in conjunction with identities, so there are restrictions on their use.
“The reason why we do not adopt the authentication with a single type of biometric information, even though it has an advantage of not having to enter private ID information, is that using only one type of biometric information has limited accuracy,” says Narishige Abe, the Project Manager in the Digital Innovation Core Unit. He has been working on neural networks since before Deep Learning made its appearance. After working on UI, UX, and image processing such as the tangible user interface, he is now playing an important role to manage the Authentication and Settlement Project as a project manager.
“At present, it is said that the maximum number of people who can be identified by palm vein authentication is about 10,000, taking the errors in authentication into account. If we want to identify a single person by using biometric authentication alone and accommodate it to a wider range of applications in real life, we need to increase this number to 1 million or 10 million units.” he says.
Thoroughly Taking the User’s Perspective to Enhance the Accuracy of the Identification on a One Million User Scale
Multimodal Biometric Authentication approach, which combines multiple biometrics to identify individuals, is the very thing Fujitsu Laboratories has focused on in this way. Facial authentication was chosen as an additional biometric authentication to integrate Fujitsu’s specialty, palm vein authentication technology. Abe explains the reason,
“What we have most cared about was the user perspective. Biometrics is expected to be used for payments as well as for entry and exit. We have considered what type of information is the best which can be naturally obtained without calling users attention to it, when people hold their palms over the sensor, for example, when shopping or passing through a gate. The best choice we have concluded is face authentication. By integrating different types of authentication, we have successfully extended the accuracy to the 1 million users scale. “
At Fujitsu Laboratories, research into facial recognition has been conducted for many years, but the challenge has been the lack of high accuracy compared to other biometrics. Afterword, due to the rapid progress in research on neural networks and AI/machine learning, the accuracy of authentication has improved dramatically.
“Key feature of our technology is not only identifying one person out of 1 million in real time, but also realizing the same convenience as the conventional palm vein authentication alone. Our technology would allow users to do hands free payment (ID-less authentication) online when they shop. It’s no longer necessary to use a smartphone or ID cards to identify ourselves. So, I think it will be very convenient and greatly improve user experience (UX).” says Abe.
Abe took advantage of the study abroad program of Fujitsu Laboratories to conduct research on the combination of biological technologies and cryptography at Stanford University. He had long wanted to make the most of this experience and knowledge, including research on UI and UX. These drove him to put Multimodal Biometrics Technology into practical use.
In March 2020, verification tests of this technology began at a Lawson store in the Fujitsu Shin-Kawasaki Technology Square. Researchers are moving forward with a new approach, in which Multimodal biometrics would be used at the gate where people pass through to enter or leave stores and at the payment.
FIDO Alliance: A non-profit standards organization established in July 2012 to standardize online authentication technologies. In recent years, the movement to adopt FIDO authentication has become active worldwide, and a more secure and convenient environment that does not depend on passwords is provided.
Striving to Achieve both “Authentication Accuracy and Processing Time”
The Authentication and Settlement Project at Fujitsu Laboratories, which is responsible for the development of Multimodal Biometric Authentication technologies consists of three groups: the Core Technology Group, which develops main technologies; the Suite Development Group, which assembles technologies developed by the Core Technologies Group into easy-to-use products; and the “Practice” Group, which verifies with business unit whether systems and services created by the suite development group are truly user-friendly. The aim is accelerating the development cycle through the cooperation of these groups toward the achievement of practical application in a short period of time.
However, integrating the two authentication technologies is never easy. In order to make it practical for everyday use, it is necessary to improve the accuracy and speed of collating each piece of biological information. Tomoaki Matsunami, an expert in such fields as image processing and pattern recognition, and a specialist in face recognition in the Core Technology Group, explains as follows.
“The biggest benefit of biometrics is convenience, but as the number of people to identify increases, processing time becomes a challenge. We use machine learning AI models for authentication, but there is a dilemma that the higher the accuracy, the longer the processing time.”
Lina Septiana from Indonesia, who joined Fujitsu Laboratories after studying image processing in a doctoral course, became a member of the Core Technology Group in May 2020. She also said, “It is the most challenging task to discover how to improve accuracy while reducing processing time. While academic research in graduate schools has focused on the work of individuals, organizational research requires teams to work together to solve problems. It’s a matter of course, but I feel the difference is refreshing. “