- The global facial recognition technology market is expected to reach $9.6 billion by 2022
- Facial recognition tech is predominantly used in security and law enforcement
- Taking over retail, advertising, healthcare, and banking in three… two… one…
Face, voice, eyes, fingerprints – these are some of our defining physical characteristics, and they’re unique to each individual. They are perfect for identification and authentication purposes, using biometric technologies. In fact, facial recognition is the most commonly used form of biometrics today, and it’s starting to play an increasingly prominent role in more and more aspects of our lives.
The global facial recognition technology market is expected to reach $9.6 billion by 2022
According to a 2016 study published by Allied Market Research, the value of the global facial recognition market is expected to reach $9.6 billion by 2022, with a compound annual growth rate (CAGR) of 21.3 per cent. Although primarily associated with security and law enforcement, thanks to recent technological advancements, facial recognition is gradually finding more and more applications in other fields as well, including retail, marketing, banking, healthcare, and many others. Facial recognition has numerous advantages over other forms of biometric identification. For instance, it’s faster and easier to deploy and implement, because it can take advantage of existing cameras and monitoring devices. Furthermore, since it doesn’t require any kind of physical interaction from the user, it’s much more practical and more hygienic than other methods.
Facial recognition technology uses a combination of image processing and artificial intelligence to identify and authenticate a person by capturing an image of their face and comparing it against faces in its database. The process is completely automated and can identify a person in a matter of seconds with nearly 99 per cent accuracy, depending on the software used. So, what exactly are the most common applications of facial recognition tech today?
Facial recognition tech is predominantly used in security and law enforcement
Facial recognition first entered the mainstream with the release of iPhone X, which put the technology right in the palms of our hands. Following Apple’s example, several other smartphone manufacturers, including Samsung and Xiaomi, incorporated the technology into their products as well. As a result, millions of people now use facial recognition tech to unlock their phones without even thinking about it.
However, security and law enforcement remain the most common applications of facial recognition technology today. It’s used to control access to various sensitive facilities around the world, such as government buildings, labs, bank vaults, or boardrooms, as well as to safeguard public events like concerts or sports tournaments. It’s also frequently used to verify the identity of people at border checks and airports and prevent known criminals and terrorists from entering or leaving the country. For example, US Customs and Border Protection (CBP) recently announced it will start using a biometric entry/exit system at the Orlando International Airport. This will require all international passengers to be scanned by a facial recognition system installed at the departure gate that will then compare the photo against the Department of Homeland Security’s database. According to CBP, the process takes less than two seconds and is 99 per cent accurate. The system is simultaneously being tested at 12 other major airports across the United States and CBP hopes to eventually make it a universal feature of every airport in the country.
Taking over retail, advertising, healthcare, and banking in three… two… one…
Retailers are also increasingly using facial recognition tech to prevent crime and identify known shoplifters or people with a history of fraud as they enter the store. According to data gathered by FaceFirst, this strategy has proven rather successful, reducing shoplifting by 34 per cent and reducing in-store violence by up to 91 per cent. In addition to preventing theft, retailers can also use facial recognition to instantly identify customers and provide them with personalised offers. A similar approach is often employed by advertisers as well, who are using facial recognition tech to detect a customer’s age and gender so as to deliver more relevant ads.
In healthcare, facial recognition technology is used to track patients’ medication adherence, accurately assess their pain levels, and detect various diseases and conditions. For instance, researchers at the National Human Genome Research Institute have found a way to use facial recognition to detect – with 96.6 per cent accuracy – a rare genetic disease called DiGeorge syndrome, which is caused by a defect in chromosome 22. Similarly, the Israeli software company FDNA has developed facial recognition software called Face2Gene that can diagnose genetic diseases and conditions like Down syndrome, Hajdu-Cheney syndrome, and meningocele syndrome. In the future, as the technology develops further, we can expect to add other diseases to this list as well.
Facial recognition technology is also increasingly used to authorise payments. This trend originated with the MasterCard Identity Check app, released in 2016, which allows customers to confirm a payment by taking a selfie with their smartphone. Similarly, the Chinese financial services corporation China UnionPay has installed facial recognition software to the majority of its ATMs in Macau. The system requires users to look into a camera for six seconds before they can withdraw cash. The Chinese e-commerce giant Alibaba is also experimenting with facial recognition payments. They recently trialled the technology at a KFC restaurant in Hangzhou where users of the digital payment platform Alipay could pay for their meals using nothing but their face.
However, despite its growing popularity, facial recognition technology has also attracted a great deal of controversy, mainly centred around privacy, reliability, and ethics. For example, a recent study published by MIT and Stanford University reveals that the gender and skin colour of the subject can have a significant impact on the software’s accuracy. While the error rate for determining the gender of light-skinned men never went beyond 0.8 per cent, it grew to an astonishing 34.7 per cent where dark-skinned women were concerned. Another major incident took place in July 2018, when Amazon’s facial recognition software called Rekognition falsely matched 28 members of the Congress with mugshots from a publicly available database of people who had been arrested for various offenses.
While there’s clearly still quite a bit of room for improvement, the benefits of facial recognition technology are simply too vast to ignore. And while facial recognition technology has traditionally predominantly been used for security and law enforcement purposes, it’s steadily finding applications in other sectors as well, including retail, advertising, banking, and healthcare. And if current trends continue, the tech will soon have permeated all areas of our lives.