FB pixel

Apple patents for a secure transition of enrolling biometric templates with every software update

Apple patents for a secure transition of enrolling biometric templates with every software update
 

Apple has potentially addressed security concerns when it comes to safeguarding sensitive biometric data, such as facial images or fingerprints, by storing processed data that are derived from raw data templates. This measure minimizes the risk of exposing biometric information in the event of a device breach.

Additionally, software updates present a challenge as pre-existing enrollment biometric data may not be compatible with the new version. However, Apple has proposed a solution that allows for a transition of the enrollment templates from the old algorithm to the updated one.

On Tuesday, the U.S. patent office granted Apple patent 11935327, “On the Fly Enrollment for Facial Recognition.” It describes an enrollment process that eliminates the re-enrollment of biometric information with every software update.

Although the existing enrollment data may not be compatible with the latest software version, the software updates often come with improvements to the algorithms and models, including neural networks that can assess and compare biometric data.

These techniques may include algorithms that can update templates or transform them into a format that works with the new software version, all while maintaining the security and accuracy of the authentication process.

During a software update, the device utilizes two facial recognition processes simultaneously: the current neural network and a new neural network. The current neural network manages authentication tasks, while the new neural network operates in a virtual mode, running in the background without interfering with the authentication process.

As time progresses, the device gathers data on the performance of both the current and new neural networks in recognizing biometric information. This evaluation assesses accuracy, speed, and security.

By analyzing performance, it can be determined whether to switch operational control from the current neural network to the new one. It’s important to transition between models gradually to maintain user trust and system integrity.

Related Posts

Article Topics

 |   |   |   |   | 

Latest Biometrics News

 

Biometrics disrupting the future of movement, on and offline

Biometrics are disrupting different areas of life, from how people interact with governments for basic services to the esoteric world…

 

Alexa, sue Amazon: tech giant faces class action over voice recordings

Users of Amazon’s Alexa are clear to pursue a class action over allegedly illegal recordings of private conversations. In Seattle,…

 

Epic Games provides Yoti facial age estimation to Bluesky for UK users

Social media platform Bluesky has selected Epic Games’ software, including biometrics-based age estimation from Yoti, to ensure its compliance with…

 

RealSense targets robotics, 3D facial recognition security with $50M in hand

RealSense has cut the cord tying it to Intel Corp, where the 3D camera company was born, with $50 million…

 

Will Congress reaffirm US cyber threat sharing framework before it’s too late?

As the September 30 expiration date for the Cybersecurity Information Sharing Act of 2015 (CISA 2015) rapidly approaches, Congress faces…

 

World pauses German operations for Orb update amid regulatory faceoff

World is facing a potential cease-and-desist order in the Philippines, and has put its iris scanning stations on hold in…

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Biometric Market Analysis

Most Viewed This Week

Featured Company

Biometrics Insight, Opinion

Digital ID In-Depth

Biometrics White Papers

Biometrics Events