walking like a bomber

[Eugen Leitl]

http://www.techreview.com/printer_friendly_article.aspx?id=18072

Wednesday, January 17, 2007
Walking like a Bomber
New strides in radar and gait-analysis software show that it's possible to
detect when someone is carrying a bomb well before he or she reaches a
security checkpoint.
By Karen Nitkin

In November 2005, three suicide bombers walked into three hotels in Jordan and
blew themselves up, killing 63 and injuring more than 100. While the world is
alert to such deadly threats, the challenge remains: how to detect approaching
suicide bombers from a safe distance. X-ray machines can obviously see a
concealed bomb, but they are dangerous to humans--and a bomber could detonate
himself and kill people at the checkpoint. Video surveillance can help, but it
requires personnel trained to scan crowds and pick out suspicious
individuals.

A new radar-imaging technology expected to reach market later this year could
solve the problem by directing low-power radar beams at people--who can be 50
yards or more away--and analyzing reflected radar returns to reveal concealed
objects. And early research indicates that this method could one day be
augmented with video-analysis software that spots bombers by discerning subtle
differences in gait that occur when people carry heavy objects.

Virginia-based SET Corporation is developing both approaches for its
CounterBomber, a system nearing commercialization that detects suicide-bomber
suspects from a safe distance, says Thomas Burns, CEO of the company, which
was founded four years ago by scientists from the Defense Advanced Research
Projects Agency. Customers might include airports and military bases, he says.
The device could be ready for sale by the fall of 2007.

The first generation of the CounterBomber works by continuously steering a
low-power radar beam toward the moving subject. The radar then repeatedly
"interrogates" the subject. "The characteristics of the reflected radar beam
are affected by weapons hidden beneath the clothing," Burns says. Signal
processing software can detect those weapons or bombs without creating an
under-the-clothes image that could violate the person's privacy, he says.

And this technology is helped by novel technology that tracks the
subject--thereby enabling the radar to be continuously aimed at the moving
person. Software developed by Rama Chellappa, a professor in the department of
electrical and computer engineering and a member of the University of
Maryland's Institute for Advanced Computer Studies, uses a form of "gait
recognition" to do this. It notes a person's walking style and physical
attributes such as height, then uses those features to follow individuals as
they move and locate them again even after they've been obscured by poles or
other objects. "Rama's technology in its most basic form currently allows us
to track the people more effectively, especially in crowds," Burns says.

But the next generation of Chellappa's technology could extend the role of
gait recognition. In early-stage research, he has shown that he can analyze
the joint movements of a walking person and tell whether those movements are
anomalous and possibly consistent with carrying heavy objects--and even
whether the person has just deposited something on the ground.

This work is at an early stage. Chellappa has created a model of human
movement based on the movements of 11 joints--including the knee, elbow, and
hip--and established a database of normal movements for a variety of body
types traveling at a variety of speeds. This forms a database of the normal
range of human movements, against which videos of a walking person can be
compared.

In a November demonstration at an army research conference in Orlando, FL,
Chellappa showed that his system could detect someone who had just
surreptitiously deposited an object on the ground simply by noting changes in
the way the person walked before and after dropping the object. And he is now
developing software able to detect the gait of people who have a 15-pound
object attached to their legs.
"We have clearly made a link between humans carrying things with them and the
corresponding changes in their walking pattern," Chellappa says. "We see
differences in the way people walk when they strap even 15 pounds to their
ankle, but it's a very subtle thing." He concedes that the work is
preliminary--and that the problem of detecting extra weight on a torso is a
research challenge--but he adds, "I believe it's a reasonable way to approach
it."

His work represents a new direction for the field of human movement
signatures, says Alex Vasilescu, a research scientist at MIT's Media Lab. Some
gait-recognition research has shown the potential for early detection of
diseases like Parkinson's. And several research groups are working on
developing a way to take a person's "gait fingerprint." This could allow a
video system to identify that person based on previously stored information.
But Chellappa's technology requires no previous information about an
individual. "It's very relevant to our times," Vasilescu says. "I would like
to know if someone is carrying a concealed weapon, and we'll worry about who
that person is afterwards."

What's really novel in this research is that rather than searching for a gait
fingerprint, the technology searches for suspicious activities, says Thomas
McKenna, project manager at the Office of Naval Research, which funded
Chellappa's work. "It's a new way of using surveillance that looks at
activities, instead of looking for people," he says.

The first version of the CounterBomber to reach market won't use gait
recognition to determine whether someone is threatening. Rather, this first
version uses only the reflected radar beam to make the determination. But the
next version of the technology could include gait recognition as a way to help
identify suspicious activity. "By incorporating Rama's full gait-recognition
technology in the next generation of our system, we will be able to combine
evidence both from the radar and the video sensors to improve our
discrimination performance," Burns says.


--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE

[demime 1.01d removed an attachment of type application/pgp-signature which had a name of signature.asc]