You might associate biometrics with the fingerprint scanner on your smartphone, or the face camera at the airport, but in truth, biometrics comprises a wide range of modalities.
Every biometric modality works along the same basic principle: capture and measure specific biological (anatomical and physiological) or behavioral characteristics, analyze them to create and store a digital representation called a template, then match new captures against the template to verify identity. Where they differ is in what’s captured, how it’s captured, and how easy and accurate the process is.
Friction ridge biometrics identifies, stores, and compares the complex patterns of the friction ridges found on the fingers or toes, the soles of the feet, and the palms of the hand. Formed in the womb, these ridges are unique to every person and are permanent – two key features of any effective biometric. The technology to capture, analyze, and store friction ridge patterns is already familiar and mature; the capture process is fast and very accurate.
Facial biometrics works by analyzing and comparing facial features and landmarks, including the relative position, size, and shape of the eyes, nose, mouth, and any marks or patterns on the skin. The biometric data can be captured in 2D or 3D formats using visible light or a combination of visible and infra-red light. Facial systems tend to be fast, easy to use, and flexible, and they can capture or match biometric data extracted from a video or photo.
Ocular biometrics works using patterns in the iris, the blood vessels of the sclera (the white outer layer of the eyeball), periocular features, or the patterns of the retina (the light-sensitive tissue at the back of the eye). The process is convenient and non-intrusive, requiring little more than having the subject look into a camera outfitted with a specialized light source. Iris biometrics also offer another advantage: they can be used in place of fingerprinting if digits are missing or fingerprints are worn-down or thin.
Voice biometrics models the subject’s speech, using qualities such as pitch and cadence resulting from the unique length and shape of the vocal tract for each person. Voice systems can be extremely convenient–you can capture, process, and match a voice over the phone, with or without using specific words or phrases–and there’s potential to combine the biometrics with spoken passwords for added security.
DNA biometrics uses DNA captured from hair, nails, body fluids or, more commonly for identification purposes, buccal swab from the inside of the cheek. DNA-based systems are widely seen as the gold standard for biometrics, given the unique and highly personal nature of DNA. At the same time, however, they are also slower, more expensive, and more intrusive than other systems. New breakthroughs in technology will help them become faster, less expensive, and more convenient.
Electrocardiogram (ECG) biometrics works by measuring the pattern of your heartbeat. This pattern is defined by a range of specific factors, including the heart’s size, shape, and position in the body, making it unique to each person. What’s more, because there’s no easy way to measure ECG without close contact, it’s difficult to clone or replicate. Wrist-based ECG systems can be extremely convenient for continuous authentication.
The Bottom Line
No one modality is necessarily the best, so one of the key tasks in biometric security is matching the right technology to the specific application, whether that’s unlocking a secure laptop via facial recognition or using iris recognition in border controls. For some high-security applications, using multiple biometrics or combining biometrics with other factors could provide even more secure forms of authentication.
Stuart Andrews has been writing about technology and enterprise IT for 20+ years, writing for many of the UK’s best-known newspapers, magazines, and websites.