[isml] Facial-recognition tech has people pegged (fwd)

[Eugene Leitl]

-- Eugen* Leitl <a href="http://www.lrz.de/~ui22204/">leitl</a>
______________________________________________________________
ICBMTO  : N48 10'07'' E011 33'53'' http://www.lrz.de/~ui22204
57F9CFD3: ED90 0433 EB74 E4A9 537F CFF5 86E7 629B 57F9 CFD3

---------- Forwarded message ----------
Date: Wed, 18 Jul 2001 12:51:40 -0700
From: DS2000 <ds2000 at mediaone.net>
Reply-To: isml at yahoogroups.com
To: isml <isml at yahoogroups.com>
Subject: [isml] Facial-recognition tech has people pegged

>From CNN,
http://www.cnn.com/2001/TECH/ptech/07/17/face.time.idg/index.html
-
Facial-recognition tech has people pegged

July 17, 2001 Posted: 1:01 p.m. EDT (1701 GMT)

By Emelie Rutherford

(IDG) -- Forget ID badges, passwords and access cards. Pretty soon, to get
in and out of your office you might start using something you can't forget
or misplace: your face.

Once the stuff of science fiction, facial recognition technology has started
to appear in real-life buildings and public places. Setups consist of
cameras that capture images of people who pose or simply walk by, and
software that matches those pictures with those stored in a database.

Institutions of all kinds -- such as those that want to protect buildings or
internal networks and banks in need of greater security for ATMs -- have
recently begun to use facial recognition to verify users. Physical access
control will be the main source of revenue for biometrics companies over the
next five years, according to Marlene Bourne, a senior analyst for emerging
semiconductor applications at Scottsdale, Ariz.-based Cahners In-Stat Group.
Currently, though, it is being used most in casinos (more than 100 across
the country have facial recognition in operation) and neighborhoods (the
city of Tampa uses it in outdoor cameras to spot missing children and
lawbreakers).

Facial recognition is a technology that has been around for a while.
University scientists have been working on facial recognition for over a
decade, with financial support from the U.S. Defense Department, in an
attempt to find a technology that can spot criminals at border crossings.
Companies began commercializing the technology in the mid 90s. It made
headlines last February, when word got out that authorities used it at Super
Bowl XXXV in Tampa to search for felons and terrorists among the crowd of
100,000 spectators.

Facial recognition technology falls under the umbrella of biometrics,
technologies that identify people based on features such as faces, hands,
fingerprints and eyes. Electronic readers can be affixed to entryways,
keyboards, laptops and mobile phones. The biometrics market rose from $6.6
million in 1990 to $63 million in 1999, according to the San Jose-based U.S.
National Biometrics Test Center. And pundits say it will continue to grow
substantially, fueled by companies in need of more advanced security
precautions. Cahners In-Stat Group predicts sales of biometrics will reach
$520 million by 2006.

While other types of biometrics, such as iris scanning, are even more
accurate than facial recognition (which has a relatively low error rate;
just under 1 percent), facial recognition will probably be accepted more
widely because it is not intrusive. It does not require that the user push,
insert or click on anything. Companies often do not need to install anything
beyond the new software because most already have cameras in place and
pictures of employees on file -- making it cheaper than iris reading setups.

"Unlike other biometrics, facial recognition provides for inherent human
backup because we naturally recognize one another," says Frances Zelazney,
the director of corporate communications at Visionics, a leading biometric
developer based in Jersey City. "If the system goes down, someone can pull
out an ID with a picture as backup, something you can't do with fingerprint
devices."

How it works
Visionics' FaceIt software measures a face according to its peaks and
valleys -- such as the tip of the nose, the depth of the eye sockets --which
are known as nodal points. "While a human face has 80 nodal points," says
Zelazney, "we require only 14 to 22 to do the recognition. We concentrate on
the inner region of the face, which runs from temple to temple and just over
the lip, called the 'golden triangle.' This is the most stable because if
you grow beard, put on glasses, put on weight or age, that region tends no
to be affected, while places such as under chin would be."

FaceIt plots the relative positions of these points and comes up with a long
string of numbers, called a faceprint. The software matches faceprints in
the existing file with those of the people passing in front of the cameras.
Faceprints can also be stored on a smart card that users swipe through a
door without looking into a camera.

Visionics' main competitor, Littleton, Mass.-based Viisage Technology, has a
slightly different model. Its software compares faces to 128 archetypes it
has on record. Faces are then assigned numbers according to how they are
similar or different from these models.

The use of facial recognition technology upsets some civil libertarians, who
call covertly scanning people's faces an invasion of privacy. Soon after
Tampa installed cameras in nightlife neighborhood Ybor City in June, House
Majority Leader Dick Armey issued a statement blasting the program's
Orwellian aspects.

"The technology is blind as a bat if you're not in the database," counters
Zelazney. "It's not automatically adding people to the database. It's simply
matching faces in field-of-view against known criminals, or in the case of
access control, employees who have access. So no one's privacy is at stake,
except for the privacy of criminals and intruders."

--
Dan S


[ISML] Insane Science Mailing List

- To subscribe: http://www.onelist.com/subscribe.cgi/isml



Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/