Everything you need to know about biometric authentication in identity verification
Every day we use our biometric information – our face, fingerprints, eyes and voice – to authenticate who we are.
Deloitte found that Australians make an estimated 100 million imprints daily using smartphone fingerprint scanners. At airports, biometric systems are used at ePassport gates to compare passport photos to facial biometrics to confirm identities at border control. An Australian bank uses voice recognition to verify identities and authorise transfers through its app.
Across sectors, biometric authentication is helping businesses move forward with the confidence that they know exactly who they are doing business with.
So, how does biometric authentication work and what technologies are available to businesses looking to establish and verify a person's identity with greater accuracy?
Here's everything you need to know about biometric authentication.
What is biometric authentication?
Everyone has physical characteristics that are unique to them. These are our biometrics. They include biological features like fingerprints, facial patterns, and our voice, and behavioural features like keyboard dynamics and walking gait.
Biometric authentication is the use of these characteristics to identify individuals. Biometric data is captured through specialised sensors or technologies such as fingerprint readers, facial recognition systems, or iris recognition systems and compared with biometric data stored in a database to confirm or deny the person's identity.
Only some characteristics can be used effectively for biometric authentication. For biometric authentication to work, every individual accessing the system must possess the same feature; the feature should be sufficiently different to distinguish any two individuals and should not change significantly with time or age.
Types of biometric authentication methods for identity verification
Biometric authentication methods include facial recognition, iris scans, voice recognition, fingerprint scanning, and behavioural biometrics.
Here are the most common types of identity verification:
1. Fingerprint recognition
Fingerprint recognition is one of the most common and everyday forms of biometrics. Many of us place our fingerprints on a fingerprint reader whenever we unlock our smartphone.
Fingerprint recognition technology identifies and verifies individuals based on their unique patterns on their fingertips. Introduced in the 1960s, the first automated fingerprint system digitised and stored fingerprint data for efficient matching and retrieval by law enforcement agencies in criminal investigations.
Fingerprint recognition remains a reliable and efficient identification method because fingerprints are unique, distinguishable, and consistent over time. They can also be acquired, measured and processed with ease.
Today's fingerprint recognition systems can use various algorithms, such as ridge-based matching, along with machine learning to improve the accuracy of fingerprint recognition.
2. Iris recognition
Our eyes are incredibly unique, particularly the iris. The coloured part of the eye surrounding the pupil has distinctive colours and patterns, which makes it ideal for biometric authentication. Iris scanning uses camera technology and infrared illumination to capture a high-resolution image of the iris. The software then extracts key features from the image to create a unique template or code. Pattern-recognition techniques compare this pattern to a database of iris images to verify an individual's identity.
Many countries are exploring the advantages of iris scanning for border security. Dubai International Airport introduced iris scanning technology in 2021 amid the Covid pandemic. Passengers look at an iris scanner for a second or two while passing through immigration control. The passenger's iris patterns are encoded mathematically and sent to a central facial recognition database. The system then verifies their identity, eliminating the need for human interaction or touch.
3. Facial recognition system
One of the most common forms of biometrics is facial recognition. Facial recognition uses AI to scan the unique features of a face, determine if a real user is present and match the user's biometrics to the identity document presented.
For example, greenID FaceMatch technology uses a set of 68 facial landmarks for each image and compares them against the document to produce a reliable conclusion around how similar the two faces are and confirm a genuine presence.
Facial recognition is increasingly used in the identity verification process for onboarding customers. Its clear advantages are leading businesses to extend its use into other areas. For example, an Australian bank is trialling facial recognition to use an ATM without a debit card.
When to use biometric identification verification
When it comes to identity verification for onboarding, biometric authentication offers many benefits – especially where enhanced customer due diligence (ECDD) is required.
Document verification is the first part of the process. The software captures the document image to match it against a library of document images, checks it is genuine, and detects tampering.
But it is not enough to know the document is genuine. You need to be confident that the person presenting the document is the real owner. For that, biometrics is the solution. Facial recognition verifies that the person pictured on the document is the same person presenting the document.
That's why businesses often combine a document and biometric-based solution for KYC compliance and fraud prevention. A user provides an image of a photo ID to be checked for authenticity along with a selfie taken at the moment of onboarding. Then, a biometric authentication system matches the facial features between the printed document face (e.g. a passport photo) and a selfie or live face image captured by a smartphone, tablet or webcam. This confirms to the business and gives them the confidence that the identity document was submitted by its true owner and not by a fraudster who may have stolen it.
Biometric verification with passive liveness detection provides an extra layer of protection for organisations. Passive liveness technology detects that the customer is really there on the other end of the device and that they match the presented ID document.
Benefits of using biometric authentication for identity verification
Biometric authentication is highly secure
Fraud attempts are not only increasing in number but also becoming more sophisticated. However, biometric authentication offers a highly secure process of proving customers are who they claim to be.
Because biometric information is unique, distinctive and mostly permanent, it is almost impossible for anyone to replicate or fake. Combining liveness checks for genuine presence makes it even harder for fraudsters to gain access illegally.
Biometric authentication enables a convenient user experience
Speed and convenience matter for customers, and biometric authentication delivers on both counts.
Consumers are accustomed to biometric checks – most of us use our fingerprints to unlock our phones or have travelled through smart gates at an airport's border control. As such, integrating biometric authentication into onboarding processes is less likely to be perceived by consumers as a barrier. Our research found that three in four Australians strongly prefer using facial biometrics for identity verification, and 73% of consumers would be willing to do liveness checks .
While some biometric authentication methods are faster than others, most can be completed in seconds. For example, GBG greenID's document verification and biometric authentication process can be completed in as little as 30 seconds. For onboarding, this helps businesses deliver a convenient and secure customer experience from the outset.
Genuine presence is hard to fake
A biometric authentication system combined with liveness detection ensures that the person submitting biometric data is real and present when the data is captured. This is almost impossible to fake.
For example, GBG greenID compares a customer's facial biometric data from their selfie image with the portrait photo on their presented ID document. At the same time, it employs passive liveness checks in the background to differentiate real-time selfies taken by genuine individuals from fraudsters attempting to use pre-existing photos or videos for spoofing.
Limitations of biometric authentication
No authentication method is infallible, and biometric authentication, while highly effective, has limitations.
As with some other identity verification methods, false matches can occur. This happens when the biometric data of two different individuals is confused, for example, with twins or similar-looking siblings, or if the database used for comparison has incorrectly recorded an individual's biometric template.
To overcome this and reduce false matches, the biometric authentication system must be able to capture high-quality biometric data and regularly update biometric reference templates.
The reverse can also happen, and the biometric system falsely rejects a known identity. This may occur when an individual's biometric characteristics change, such as a marked change in appearance due to weight loss.
There are ways to limit errors. Layering with document and identity verification checks means businesses can build a more complete picture of a customer's identity and keep fraudsters out. Also, multimodal biometrics, which combines multiple biometric characteristics, such as facial and voice recognition, can enhance accuracy and reliability.
Move forward with confidence with biometric authentication
The more evidence you have that an identity is genuine, the more confident you can be doing business with that customer. By incorporating biometric authentication as part of your identity verification processes, you can deliver essential KYC compliance and security for the business while ensuring the quick and seamless experience customers want.
Find out more about how GBG's greenID leverages AI-powered facial verification and liveness detection to help enhance identity verification for your business.
Sign up for more expert insights
Hear from us when we launch new research, guides and reports.
