NOTE: This is a repost of a blog article written in September 2017. Face recognition technology is getting more attention recently, so we decided to repost articles our founders have written.
Governments all over the world are looking for ways to increase border security without making border crossing more unpleasant for both the passengers and the people responsible for security. They are experimenting with biometric facial recognition technology and it’s easy to see why. The technology has many advantages over traditional “flash pass” ID verification. Customs and Border Protection (CPB) started testing systems using facial recognition at Dulles Airport in 2015. The tests expanded to New York’s JFK Airport in 2016. Kiosks that use face-reading technology will be appearing at Ottawa International Airport soon, and British Airways is rolling out a similar system at London’s Heathrow Airport, comparing faces captured at security screenings with a separate capture at the boarding gate.
Now, a new project called Biometric Exit is ready to bring that same technology to every international airport in America. The project uses face recognition technology to identify visa holders as they leave the country. Currently, visa holders are only checked upon entering the country. So there is no easy way to know if they have stayed past their exit date and no real incentive for them to leave. Passengers traveling on a visa would have their photos taken right before boarding, to be matched with the passport-style photos provided with the visa application when they entered. If there’s no match in the system, it could be evidence that the visitor entered the country illegally. So how does it work?
How Face Recognition Technology Works
Biometric face recognition technology identifies or authenticates individuals by comparing their face against a database of known faces. The process can be broken down into 4 general steps:
- Enrollment: the person submits a photo to a database.
- Face Finding: the computer finds the face in the image.
- Template Creation: a numeric representation of the image is created based on the relative position, size and shape of facial features.
- Matching: the template is compared to templates in the database.
Biometric face matching is used to confirm an asserted identity or to identify an unknown person. Because faces are unique, this technology can easily detect when a person is using multiple identities. The same technology is also used for secure facility access control, identity theft prevention, fraud prevention and workforce management (employee identification for time clocks, etc.)
What are the advantages of face recognition?
The use of facial recognition for border control has numerous advantages, including:
Currently, TSA agents look at a driver’s license or other government issued I.D. and then look at the person presenting it to verify they are the same. This is a tedious task that is prone to failure. Humans just aren’t very good at it. Study have shown that accuracy of human face matching can be as low as 50%. Automating this step allows immigration officials to concentrate on other important activities.
Using facial recognition for traveler identification is more acceptable to travelers than interacting with an agent and presenting and ID. Generally, people surveyed said taking a photo was easier and faster. It’s also more convenient for everyone because travelers can scan their passport and proceed to the exit without receiving assistance from an immigration official.
Facial recognition technology is not perfect but it is accurate – more accurate than humans doing the same task. More accurate identification increases border security. Facial recognition biometrics also eliminates the possibility of human error in immigration processing.
Due to its passive nature, facial recognition is often favored over other forms of biometric identification such as fingerprinting for border control. Facial recognition devices used to take traveler images do not require any physical contact helping to increase acceptability among travelers.
Who’s using it now?
Probably more places than you realized. Some of these programs have been in the news, but others were not covered broadly.
- Starting in June, JetBlue will start testing facial recognition technology for flights from Boston to Aruba.
- Washington Dulles airport will use biometric exit technology on one flight daily from Washington, DC to Dubai.
- Delta is launching a pilot program in Minneapolis-St. Paul where some passengers will check their bags automatically through kiosks that use facial recognition software to identify ticketed passengers.
- Finland’s national airline Finnair and airport operator Finavia are trialing the use of facial-recognition technology to speed up airport check-ins.
- British Airways is introducing facial recognition-based border control technology that will seamlessly and quickly identify passengers at boarding gates in London’s Heathrow Airport.
- Venice Airport Authorities Plan Expansion of Biometric eGates.
- The Australian government is providing biometric screening technology for arrivals at the country’s international airports.
- Japan is preparing new legislation that will pave the way for a biometric entry and exit program for citizens traveling to and from the country’s airports.
Biometric face recognition technology can increase border security by making it easier to quickly and accurately identify travelers. Unlike other uses of biometric identification, there is a low risk that the technology will be abused because people accept and expect to be identified when they travel. Facial recognition is more accurate than human identity verification, faster and more convenient. Expect to see these programs expand rapidly.