What is Facial Recognition?

While facial recognition can also be used for authentication purposes (to verify the bearer of a passport, driver’s license, etc.), FaceFirst is an identification system that does not require the individual to carry any corroborating personal identification in order to establish his or her identity.

What makes an identification system effective?

• Uniqueness (for example, every human face is distinct from all others)
• Accuracy and speed
• Robustness of the information retrieval system
• Reliable performance (low false-accept/reject and false-match/non-match rates)
• Ease of use (plug-and-play; no training needed to operate)
• Acceptance by the public
• Security (safeguards against theft, loss or reproduction of stored data)

How facial recognition stacks up

It’s important to make a clear distinction between the relatively low performance levels of early (pre-2005) facial recognition technology and today’s top-grade systems, which take advantage of extraordinary advances in 3D algorithms. The identification rate of FaceFirst’s proprietary technology has increased from slightly more than .75 in 2002 to more than .95 in 2006, as measured in the U.S. government-sponsored Face Recognition Vendor Tests.

Finger scans have a slightly better positive-match record than facial identification. However, public databases contain 1.3 billion photos of individuals, while fingerprints sets number only several hundred million. Additionally, while everyone has a photographable face, about five percent of the population are unable (for various reasons) to leave a readable fingerprint.

Facial recognition is also the easiest biometric to use. Unlike finger and retina scans, it does not involve personal contact with individuals and does not impede traffic flow. For one-to-many environments — such as crowded transportation facilities — and perimeter surveillance, facial identification is the only feasible biometric system.

In a post-9/11 world, the public have come to accept the protective role of CCTV cameras.

Facial recognition: The FaceFirst advantage

Other facial recognition systems are completely dependent on automated systems, which require statistical parameters to be set for allowable matches. If the match level is set too low, the false-positive rate zooms upward, resulting in confrontations with innocent customers and potential privacy rights issues. If the match level is set too high, lawbreakers will slip through, undetected.

ABG solves this problem in two ways.

First, FaceFirst utilizes the world’s most highly developed facial recognition software — patent-pending technology rated number one by the 2006 Face Recognition Vendor Test, conducted by the U.S. Commerce Department’s National Institute of Standards and Technology.

Second, FaceFirst is the only facial recognition system that adds a human interface — The FaceFirst Operator Center software adds the ability for a human to verify a match when the software can’t positively detect one.

Where are facial recognition systems used?

The cities of Santa Ana, CA and Virginia Beach, FL, Logan Airport in Boston, Fresno International Airport, Panama's Tocumen International Airport, numerous U.S. embassies and more than 100 casinos are among the users of facial recognition systems.

In Europe, the London borough of Newham built a facial recognition system into their public CCTV system. The German Federal Police have implemented a facial recognition system enabling individuals to voluntarily enroll their facial information in order to move through fully automated border controls at Frankfurt Rhein-Main international airport.

Up until now, the high cost of facial recognition has limited its use to government entities and large-scale enterprises.