NOTE: This is a reprint of an article originally published in July 2017 on a different blog. We are reposting them on the Blink Identity blog because these issues are important and we want to keep our writing on these issues in one place.
There are many different biometric modalities but biometric matching always works the same way. First a person must be enrolled – the biometric is collected and source documents are reviewed to create a biometric identity. For matching, the person will present that biometric to a sensor and the stored record is compared to the one that has been collected. If they match, the identity is verified. But what biometric is being used? Let’s look at them:
A fingerprint is the pattern made by the ridges on your fingertips. The ridges form patterns that can be classified and the direction of the ridge end points and bifurcations (where they fork) are mapped. Fingerprints are the most popular biometric — which makes sense when you consider that criminals often leave fingerprints at the scene of a crime. This is not a new technology; fingerprints were used in by law enforcement in the late 1800s. Fingerprints are mentioned in a Sherlock Holmes short story in 1903. The FBI developed automated fingerprint scanners in 1975. In the late 1990s, commercial fingerprint verification products emerged. By 2006, fingerprint readers were added to many laptops, and in 2013, the iPhone 5S released with Touch ID.
Palm is similar to fingerprints; palm recognition looks at the palm’s ridges, texture, spatial attributes, and geometric characteristics. Using palm prints dates back more than 150 years. Interestingly, 30% of the prints lifted from crime scenes are of palms, not fingers.
The vein patterns in a person’s hand are unique, do not change with age, and are difficult to forge. In this biometric, near-infrared light reveals a hand’s or finger’s blood vessel patterns to create a unique biometric template. An advantage of vein biometrics is that they are below the skin level and unaffected by changes in the skin surface.
You think I’m kidding, but this is an emerging biometric. Like most human characteristics, knuckles have unique patterns of geometry and creases.
Eyes and Face Biometrics
The concept of identification with a retina was theorized in 1935 but the technology wasn’t available until 1975. In this biometric, a person is identified by the unique pattern of blood vessels at the back of their eye. All you really need to know about retina is that it isn’t used anymore. The enrollment process was difficult and sometimes uncomfortable and it has largely been replaced by iris.
In authentication with the iris biometric, the detailed structure of the iris is illuminated with near-infrared light. The texture of the iris is developed in the womb and is random so identical twins have different iris patterns, and even the left and right eyes of the same person have different patterns. An iris scan can be captured through clear contact lenses, eyeglasses, and non-mirrored sunglasses.
Face matching is composed of two parts. First the computer has to find the face – called face detection. Once a face is detected, it can be matched by comparing the overall texture of the face and location of landmark items such as nose, eyes and mouth. Face matching is particularly sensitive to changes in pose and lighting.
The voice biometric is useful for remote authentication because of easy access to phones and computer microphones. The physical structure of the throat determines acoustic patterns which are unique between individuals and can be matched.
This behavioral biometric involves analyzing the unique patterns of a person’s manner and rhythm of typing. This is not a pass/fail authentication that can be used for identification of a person but can be used to verify an expected person is behind a keyboard.
This behavioral biometric measures the unique idiosyncrasies in how a person walks. The cyclical movement of walking is an unobtrusive identifier. The emerging technology is still affected by footwear, knee injuries, terrain, fatigue, and passage of time.
This biometric looks at the unique angles and curves in each person’s ear. Characteristics include height, corresponding angles, and inner ear curve. One of the advantages of ear biometrics is that identification can take place at a distance.
As scanner technology has improved, individual sweat pores can be seen in addition to fingerprint ridges. These sweat pores are unique and can be used to match individuals. Often, sweat pores can be a key distinguishing characteristic in a latent crime scene fingerprint.