What is biometric technology?

Artificial intelligence is being merged with biometrics to help improve the security of computers and smartphones. This article explains the topic.

Biometrics is defined as the study and application of scientific and/or technical methods designed to measure, analyze and/or record unique physiological or behavioral characteristics of humans.

Biometric technology uses characteristics that are unique between people to make ourselves a means of identification/verification without having to enter strong passwords or lengthy PINs.

In fact, many of us now use biometrics in the form of fingerprints and faces.

Although biometric technology has been used by various industries for decades, modern technology has helped it gain greater public awareness. For example, many of the latest smartphones come with fingerprint scanners and/or facial recognition to unlock the device.

Compared to so-called token-based (e.g. keys, ID cards, driving licenses) and knowledge-based (e.g. PIN codes, passwords) access control methods, biometric features are more difficult to crack, steal or forge.

This is one of the reasons why biometrics are often favored for advanced security entry (e.g. government/military buildings), accessing sensitive data/information, and preventing fraud or theft.

The characteristics used by biometrics/authentication are primarily permanent, which provides convenience - you can't simply forget or accidentally leave them somewhere in your home.

However, the collection, storage and processing of biometric data, particularly in consumer technology, often raises concerns about personal privacy, security and identity protection.

There are many types of biometric characteristics in use today, each with different ways of collecting, measuring, evaluating, and applying them. Physiological characteristics used in biometrics are related to the shape and/or composition of the body. Some examples are (but are not limited to):

  • DNA
  • Fingerprint/palm print
  • iris/retina
  • Face
  • vein geometry
  • smell/smell

Behavioral characteristics used in biometrics (sometimes called actometrics) relate to unique patterns exhibited through behavior . Some examples are (but are not limited to):

  • voice
  • gait
  • sign
  • button
  • heartbeat

Features are selected because of specific factors that make them suitable for biometric measurement and identification/authentication. The seven factors are:

  • Universal – everyone must have it.
  • Unique – There should be enough difference to distinguish different individuals.
  • Persistence – Resistance to change over time (i.e. how well it resists aging).
  • Collectability – Ease of acquisition and measurement.
  • Performance – Speed ​​and accuracy of matching.
  • Evasion – The ease with which something can be counterfeited or imitated.
  • Acceptability – How open people are to a particular biometric technology/process (i.e. easier and less invasive technologies, such as fingerprint scanners in smartphones, tend to be more widely accepted).

These factors can also help determine whether one biometric solution is better suited for a situation than another. But cost and the entire collection process are also taken into consideration. For example, fingerprint and face scanners are small, cheap, fast, and easy to implement in mobile devices. That’s why smartphones use these features instead of hardware to analyze body odor or vein geometry!

Biometrics/authentication starts with the collection process. This requires sensors designed to capture specific biometric data. Many iPhone users are probably familiar with setting up Touch ID, where they have to put their finger on the Touch ID sensor over and over again.

The accuracy and reliability of the equipment/techniques used for collection help maintain higher performance and lower error rates in subsequent steps (i.e., matching). Basically, new technologies/discoveries help improve processes through better hardware.

Certain types of biometric sensors and/or collection processes are more common and pervasive in daily life than others (even if not related to identification/authentication). consider:

  • Forensic Science: Law enforcement agencies regularly collect fingerprints, DNA samples (hair, blood, saliva, etc.), video surveillance (face/gait recognition), handwriting/signatures, and audio recordings (speaker identification) to help establish crime scenes and identify identities. personal. This process is often depicted (i.e. dramatized with varying degrees of practical realism) in movies and television shows. You can even buy forensic science toys for aspiring detectives.
  • Computer Security: Fingerprint scanners are an evolving security feature that will be integrated into mobile devices - these scanners have been available for desktops/laptops (both integrated and as separate units) for years. Facial recognition technology, found in smartphones such as the Apple iPhone X with Face ID or any Android phone using Google Smart Lock, can perform security operations (usually unlocking) in place of or in addition to a fingerprint scanner.
  • Medical: Many annual wellness exams include digital retinal imaging as an (optional) enhancement to a comprehensive eye exam. Photos of the inside of the eye can help doctors screen for eye diseases/conditions. There are also genetic tests, which doctors use to help determine an individual's risk and outlook for developing a genetic disease/condition. Paternity tests are also common (often a recurring theme on some daytime talk shows).
  • Home Entertainment/Automation: Speech recognition (as opposed to speaker recognition , which is used by forensics staff to identify individuals through speech patterns) has been around for quite some time. It is mainly used in word recognition, such as speech-to-text, language translation and device control. If you've ever had a conversation with Apple's Siri, Amazon's Alexa, Android's Google Now, and/or Microsoft's Cortana, you've already experienced how entertaining speech recognition can be. Many smart home devices can also be automated with voice activation.
  • Purchases/Contracts: If you have ever paid with a credit card and/or entered into an agreement with a person/entity (e.g. ID, bank check, medical/insurance, title/deed, will, lease, etc.) you may have had to sign your name. Such signatures can be examined to help determine identity and/or forgery—trained professionals can tell the difference between natural variations in one person's handwriting and those that indicate a completely different authorship.

Once a sensor (or sensors) captures a biometric sample, computer algorithms analyze the information. These algorithms are programmed to recognize and extract certain aspects and/or patterns of features (such as the ridges and valleys of a fingerprint, the network of blood vessels in the retina, the complex markings of the iris, the pitch and style/rhythm of a voice, etc.) and will typically Data is converted to numerical format/template.

Numeric format makes information easier to analyze/compare with other information. Good security practices will involve encryption and secure storage of all digital data/templates.

Next, the processed information is passed to a matching algorithm, which compares the input to one (i.e., authentication) or multiple (i.e., identification) entries held in the system database.

Matching involves a scoring process that calculates similarity, error (e.g., imperfections in the collection process), natural variance (i.e., some human characteristics undergo subtle changes over time), and so on. If the score passes the matching minimum score, the system successfully identifies/verifies the individual.

When it comes to biometrics, the terms "identity" and "authentication" are often confused with each other. However, each is actually asking a slightly different but distinct question.

Biometrics Want to Know Who You Are - The one-to-many matching process compares the biometric data input to all other entries in the database. For example, an unknown fingerprint found at a crime scene will be processed to identify who it belongs to.

Biometric authentication wants to know if you are who you claim to be - the one-to-one matching process compares the biometric data entered to an entry in a database (usually one you have previously registered for reference).

For example, when you use a fingerprint scanner to unlock your smartphone, it checks to make sure you are indeed the authorized owner of the device.

  • Biometric screening refers to the clinical assessment of a person's physical characteristics and health status, providing a snapshot of their current health status. Height, weight, BMI, blood pressure, etc. are usually assessed. These are usually carried out by the employer or throughout the immigration process, although they can be used in other circumstances as well.

  • The process varies, but after completing the biometric appointment and submitting the accompanying documents, it usually takes 6 to 10 months to process before you receive your green card.