Between 2020 and 2025, the global biometric technology market is expected to grow at a compound annual growth rate (CAGR) of 10.5%, reaching a total value of $8.5 billion. With this steep upturn, the investment in research and development in this sector is also skyrocketing. Airports are ramping up their investments, the interest from venture capitalists (VCs) is maturing and tech partnerships are popping up like daisies – the global biometrics ecosystem is sizzling.
2022 will be a watershed year for the industry, bringing significant advancements in the use of biometrics, its security, customer experience and scalability.
Let’s explore the trends you will soon see within the biometrics technology industry.
1. Edge Computing
The latest advances in GPU processors have inspired the creation of dedicated computing platforms that are extremely useful for facial recognition. These platforms are able to process video data streams on edge – without the need to transfer large data streams to central servers.
Edge devices such as NVIDIA Jetson or Blaize can preprocess video streams on their own. The edge devices provide face recognition, extraction and template creation and only send the results to a central server for comparison. This reduces the required throughput of the local data connection from as much as tens of megabits by several orders of magnitude.
Moreover, edge devices also work with existing cameras, cutting down the costs of upgrades. By deploying edge devices or mixing and matching them depending on the required setup, users can operate more cameras simultaneously or reduce the bandwidth, leading to better cloud operability.
2. Biometrics in the Cloud
With the reduction in required bandwidth, processing-intensive tasks can now be uploaded to the cloud instead of a dedicated local server with only a small performance hit – an unthinkable task for real-time facial recognition just a few years ago. This immense progress opens a range of new possibilities, including remote monitoring of several premises at once.
3. Accuracy in Image Fragments
The combination of neural networks, high-resolution fingerprint sensors and small area detectors led to breakthroughs in accurately verifying fingerprints – even with minuscule fragments. Although this is critical to improving insight into criminal investigations, the potential use of this technology is much more diverse. It works for small sensors in smartphones to fingerprint sensors in biometric credit cards. With credit cards, identification systems can verify the holder by using a small section of a fingerprint and won’t even need a battery because the technology is powered by a minimal current at a point-of-sale terminal. Overall, the use of biometric fingerprints in the banking industry will help prevent credit card and financial fraud.
4. Multifactor Authentication
As cyberattacks are skyrocketing, it’s no surprise that IT professionals at enterprises such as Google are deploying multifactor authentication (MFA) for user logins to improve their security infrastructure.
MFA methods add an additional layer of security because they work with several authentication factors that a user has to provide with each login attempt. These factors are usually something a user has (a phone token or device such as a USB key), something the user knows (a code) or something the user is (biometric identity). Biometric technology is becoming the most sought-after element, replacing easy-to-hack passwords and standard single-factor authentication.
MFA has often been used to protect highly sensitive data – such as your primary email, financial accounts and health records – but the demand for enhancing privacy and restricting permissions is spreading across all areas, standardizing MFA.
5. Improving Accuracy in Specific Segments
Despite its fast evolution, in some segments, biometrics still faces hurdles for producing reliable results, for example, with children’s fingerprints and faces. The smaller the fingerprints, the harder it is for algorithms to properly detect their minutiae. Several emerging technologies now focus on these specific use cases and work on proper identification – an extremely relevant asset when fighting child trafficking and abuse cases.
6. Seamless Biometrics
Both the recent advances in edge computing and new ways of detecting an identity help develop new approaches for seamless identification. For example, new sets of algorithms can detect a person walking to a turnstile, give access (if it’s an authorized entry), call an elevator to the proper floor and open only those offices the given person will use.
Facilities such as airports are top of the list for this technology. The detection of unwanted persons (e.g., known criminals) and other security features can provide faster movement through airports and lead to better overall security. But the use cases don’t end there. In hospitals, biometrics can identify unconscious patients, warn doctors about patients’ allergies or provide their insurance status.
Therefore, seamless movement can be lifesaving.
7. Highlighting Ethics and Mitigating Bias
In light of the recent suspect cases such as ID.me – the IRS fiasco – or the Clearview case, companies will become more interested in the quality of biometric algorithms when deciding on a provider. Those considered biased or unethically created will have a hard time finding business partners for fear of negative reputation backlash, especially in fields in which reputation is important for business (e.g., finance). This will start a positive backlash against alleged scammers or careless providers – improving the standards of the whole industry as a result.
8. Biometrics Beyond Identification and Verification
Although biometrics have been developed first and foremost with accurate identification and verification in mind, they can and will serve many other purposes. Just think of age verification at self-checkouts in the supermarket or online shopping for alcohol. All you need is a simple scan of an ID, comparing the document photo with the selfie, and automatically reading the birth date.
In addition, body language and motion detection is valuable for quickly identifying alarming behavior on public surveillance cameras or dangerous accidents, such as heavy machinery colliding with a person in a factory. Biometrics combined with object detection can, for example, ensure that construction workers wear a hard hat and reflective vest when entering a hazardous area.
The year 2022 has a lot to offer in the biometric technology market. Not only are the use cases increasing, but the performance, reliability and ethical validity of biometric technologies are also continually improving.
Nevertheless, potential users should not blindly accept every offer and install biometric identification without hesitation. As with any technology, you want to be well informed, put the provider and its algorithms to the test and choose a low-risk, high-reliability product.
