Biomteric trendsGrowth of facial recognition biometrics, I

After a driver sits for a photo at the Illinois Secretary of State office to renew a license, officials use facial-recognition technology to give the resulting image a close look. First, state officials verify that the face matches the images portrayed on previous licenses issued under the driver’s name. The second, more extensive run-through determines if the same face appears on other Illinois driver’s licenses with different names. Washington Technology’s Alice Lipowicz writes that since starting the program in 1999, the state has uncovered more than 5,000 cases of multiple identity fraud, according to Beth Langen, policy and program division administrator at the Illinois Secretary of State office. The state pays Digimarc Corp. about 25 cents per license for the service, she said. “We are very pleased. It is a fraud for which we have no other tool” to combat, Langen said.

About 40 percent of the nation’s drivers will undergo such facial-recognition database checks when they renew their licenses in twenty states. It is just one indication that after years of ups and downs, facial-recognition technology in government agencies is gaining momentum on several fronts. Facial-image-matching applications have been available for more than a decade but are just beginning to attain widespread use in government. Using captured facial images which are adjusted for lighting, the technology extracts data from the image — such as the length of a nose or a jaw line — and uses an algorithm to compare the data from one image to other images. Facial recognition got off to a bad start when tested at the Super Bowl in Tampa, Florida, in 2001. Surveillance images of faces from the crowd generated so many false positives that the test was deemed a failure. Experts concede there still are high error rates if facial recognition is applied to images taken under less-than-ideal conditions. That type of application also spurs the greatest concern about privacy and civil rights violations.

Now, however, facial recognition is considered reliable in environments in which the lighting, facial expression, angle of the head, and distance of the subject from the camera can be controlled, and interference from hats, sunglasses, and such can be minimized. The most recent test results announced in March 2007 by the National Institute of Standards and Technology (NIST) showed error rates of 1 percent or less, a huge improvement compared with previous tests. Spending for 2008 on contracts related to facial recognition is estimated at $400 million, said Peter Cheesman, a spokesman at International Biometric Group (IBG), a New York-based consulting firm. That includes $254 million for civilian agencies, $68 million for law enforcement, and about $75 million for surveillance and access control, he said. State driver’s license bureaus are in the forefront. The twenty or so state motor vehicle departments that have facial-recognition systems or are in the process of implementing them typically perform one-to-one and one-to-many matches within their states.

Growth in such applications is continuing, driven by concerns about identity theft and fraud. Along with Colorado, Illinois, Iowa, Kentucky, Wisconsin, Washington, and many others, Oregon is the latest state to install facial recognition. “Doing facial matching in state motor vehicle departments is acceptable, logical and inexpensive. More states will move toward it,” said Raj Nanavati, partner at IBG.

On Monday: A broader perspective on facial recogntition biometrics