We recently finished our series on artificial intelligence and its multitude of uses in physical security. A natural segue is moving into biometrics and how advances in AI are making its use for access control, identification and authentication increasingly popular.
The dictionary definition of biometrics is, “The measurement and analysis of unique physical or behavioral characteristics especially as a means of verifying personal identity.” Unique physical features can mean anything from fingerprints and facial features to speech and keystroke patterns.
Biometrics have been around for a long time. There’s evidence cave dwellers used handprints as a form of “signing” their paintings on cave walls and as early as the late 14th century Chinese parents were using palm and foot prints to distinguish young children from each other. In the late 1800’s, law enforcement and courts began to see fingerprints as a useful forensic tool for solving crimes.
It wasn’t until the development of computers with large, easily searchable databases that biometrics begin to take off. Advanced computer technology allowed the collection of large amounts of biometric information that could be easily accessed and analyzed.
Other things that helped boost the popularity of biometrics include:
- Passwords – Passwords are not only difficult for people to remember, but unlike biometrics they can also be easily shared or hacked. According to a recent survey, passwords account for 81 percent of data breaches.
- Biometric-enabled mobile devices – The proliferation of biometrics in smartphones and mobile devices has been a real driver for the biometric market. A recent study by Juniper Networks predicts 80 percent of all smartphones – or 5 billion units – will have some form of biometric hardware by 2023.
Here are the big three biometric technologies used most frequently today:
- Fingerprint – This biometric modality was given a big leg up by Apple and other mobile smartphone manufacturers.
- Facial recognition – There has been a big jump in this technology. Again, mobile device makers, such as Google's Pixel 2, have adopted the technology while the use of AI and machine learning has made it more reliable.
- Iris recognition technology – Considered the most accurate of the biometric technologies, iris recognition is non-invasive, non-contact and very fast.
While biometric fingerprint technology is the most widely used many experts believe there will be rapid growth in both facial recognition and iris recognition as costs continue to come down.
There are many other biometric technologies based on physical traits including palm prints, vein patterns and heartbeat recognition. On the behavior side, biometrics can identify people by their voice, gait or key strokes on a keyboard.
Biometrics are used to prevent hacking and data breaches, but in physical security they are being used for access control, authentication and verification. Unlike keys, ID’s, proximity cards and passwords biometrics can’t be stolen or lost making them a valuable tool for physical security. And when you combine physical security benefits with digital security features you get an end-to-end technology that can work throughout the organization giving people secure access to both physical and digital assets.
In later blog posts we’ll talk more about the specific biometric technologies, how they work and the pros and cons of each.