Diverse automotive biometric and computer vision systems continue to evolve

Honda is working on a keyless system for starting electric scooters based on biometric facial recognition embedded in a helmet. A patent filing published by the U.S Patent and Trademark Office and spotted by RideApart describes a system to prevent the theft or loss of keys, as well as preventing a running scooter from being stolen by shutting off the engine when the rider steps off. The helmet and scooter would likely be sold together, with additional users possibly required to purchase a separate helmet.

RideApart notes that the patent filings do not appear to depict an existing Honda scooter, so the timeline for possible availability is unknown, but guaranteed helmet usage and theft prevention seem like potential selling points.

Suburu, meanwhile, has announced that the 2019 Forester, which features facial recognition, has been recognized among the 10 Best Family Cars of 2019 by the publication Parents. Parents lauds the Forester’s modern makeover, and Suburu says the DriverFocus Distraction Mitigation System, which provides audio and visual warnings in the case of driver fatigue or distraction, uses a dedicated infrared camera with facial recognition technology.

Autonomous vehicle sensor development rolls on

NVIDIA has announced its new NVIDIA DRIVE AP2X level 2-plus automated driving solution with biometric facial recognition capabilities at the GPU Technology Conference. The system consists of the DRIVE AutoPilot software, and DRIVE AGX and validation tools, according to a blog post on the announcement. The facial recognition system is powered by a deep neural network (DNN), and can be used to open the car door, start the engine, and make adjustments to seating and cabin conditions. The system’s ClearSightNet feature also allows the vehicle to detect camera blindness, such as when direct sunlight or other weather conditions prevent a sensor from working properly.

The autonomous driving capabilities of the Cadillac Super Cruise system depend on infrared cameras which sometimes shut off when hit with direct sunlight, Driving reports. Cadillac’s system relies on facial recognition to monitor driver attention, unlike some others that use manual inputs like the pressure of hands on the steering wheel.

Other companies are working on sensors to capture brain wave or facial expression data from drivers. The Globe and Mail reports that Freer Logic is planning to add its neuro-monitoring technology to vehicles in 2020 through partnerships with automakers and parts suppliers. Affectiva is also looking to 2020 to have its dashboard or rear-view mirror-mounted camera deployed to vehicles to detect fatigue, distraction, and other cues from facial images.

Both systems are pitched as helpful for level three autonomous driving systems, which would likely involve control being passed between the driver and the automated system.

Autonomous vehicles could free drivers up to take more advantage of personalized services relaying on identity, and auto makers are pursuing biometric capabilities to enable those kinds of services, Synaptics VP of Marketing Godfrey Cheng told Biometric Update in October.

Article Topics

automotive biometrics  |  computer vision  |  facial recognition  |  Nvidia  |  patents  |  sensors