Fujitsu unveils multimodal biometric authentication technology for contactless retail
Fujitsu Laboratories has unveiled new multi-factor biometric authentication technology designed to deliver a contactless shopping experience for the post-COVID-19 “New Normal” era.
The new solution combines two different forms of biometric authentication, matching facial verification (even when users are wearing masks) with palm recognition for a reported accuracy of more than 99 percent.
Fujitsu has been working steadily on biometric technologies in the past few months.
The company says it recently ranked 6th overall among 147 participating global vendors and 1st amongst Japanese vendors in NIST’s original test of face biometrics with masked faces.
To update its face recognition tools to identify people wearing masks, the company now trained its algorithms to generate a composite image combining the mask with the face image and training the model, thus achieving great filtering accuracy, according to the announcement. Fujitsu also announced the development of a liveness detection technology for face authentication last year.
Fujitsu’s palm vein authentication sensor platform has also been improved by adding a palm-shaped light, in which color and light pattern change according to the distance of the hand. This helps users to understand more intuitively how to position their hand on the sensor to perform biometric authentication.
The new, contactless solution combining face and palm recognition is being trialed this month at a concept cashless Lawson convenience store at the Fujitsu Technology Square office in Shin-Kawasaki, Japan.
The company aims at commercializing the technology by the end of the year.
Earlier this month, Fujitsu also developed a model for higher accuracy in complex behavior recognition.