Ongoing research into large-scale, video-enabled wireless surveillance networks in the department of Electrical and Computer Engineering at Virginia Tech ironically may one day help deter the very type of shooting rampages that occurred in the university's engineering building April 16.
Wireless surveillance networks provide accurate, real-time visual data from the field.
We'll never know whether such advanced video surveillance technologies could have mitigated the shootings, but one thing we do know is the trend is away from traditional analog closed circuit television (CCTV) systems and toward wireless, digital, Internet Protocol-based (IP) video surveillance.
"Security professionals are increasingly concerned with new threats, such as WMDs and explosives devices such as truck bombs, and thus they're applying the most appropriate solutions for thwarting those threats," says Jeff Penny, vice president of ICx Technologies' Surveillance Division located in Washington, D.C.Industry swing
The shift toward wireless IP systems is simply because more can be done with IP than traditional analog systems.
"IP security solutions are scalable, extremely flexible, more easily integrated into IT networks, and allow for massive customization to specific security and operational needs," says Craig Chambers, CEO of Cernium in Reston, Virginia.
This has a clear impact on the delivery of video surveillance technology in many sectors, particularly law enforcement, which by its nature must receive and react to information from remote locations.
New surveillance solutions also now include technologies such as 3D and sweeping panoramic views generated by computers that combine multiple camera views into a single picture.
There are several other advantages of digital over traditional analog systems. One is clarity of image, crucial in security.
"Analog cameras typically produce blurry, grainy images due to their inability to handle as many lines per image as IP cameras," says CEO Ralf Hinkel of Mobotix in New York, New York. Standard cameras record at 288 lines, whereas some IP cameras record at 960.
"Remember the foggy image of Mohammed Atta going through airport security on 9/11?" Hinkel asks. "That was an analog image."The eyes of excess
The paradigm shift in surveillance does not come without controversy. The proliferation of video surveillance raises fears that freedoms of speech and association are being eroded.
Chicago and New York already utilize more than 3,000 cameras. The New York Times estimates there are an estimated 4.2 million CCTV cameras in Britain — one for every 14 citizens. London has approximately 500,000 cameras in operation, a legacy of their years of assault by the Irish Republican Army.
"Surveillance cameras inflict a kind of injury, whether or not they reduce crime," says New Jersey attorney Grayson Barber, a First Amendment litigator and privacy advocate. "They constitute a form of government-sponsored intrusion on personal conduct, even when that conduct is perfectly legal."
Barber maintains that those who say there is no expectation of privacy in public places are wrong. He contends that just because you're in a public place, the government cannot intrude on your business unless it has a good reason.
"We shouldn't feel safer that our government is treating all citizens as potential suspects by keeping them all under surveillance," Barber adds. "Instead of saying 'Follow that car!,' they're saying 'Follow all cars!' It doesn't compute."
Francoise Gilbert, managing director of IT Law Group in Palo Alto, California, says surveillance technology could be abused for purposes unrelated to law enforcement, such as to create secret profiles of individuals.
"Suppose mayor Smith's wife is filmed entering an abortion clinic, or Rev. Jones is recorded attending an AA meeting," he gives as an example. In the wrong hands sensitive information could be used illegitimately for political gain, to prevent employment opportunity or to deny insurance access.
"Clear procedural guidelines and legislation are needed to address the usage, sharing and retention of video obtained in public places," Gilbert comments.
The surveillance cart may be before the legal horse, but it hasn't stopped deployment of new video technology. Implementation of video networks often makes financial sense through consolidation of security functions.
Security officer positions could be eliminated, for instance, by utilizing consolidating technologies that link feeds from separate building systems into one corporate-park system. Here a single controller monitors the entire complex, instead of each building having its own security staff.
"Technology now permits the consolidation of all video feeds, while combining them with devices such as motion sensors," says Santo Scribani, CEO of ABM Security in Houston, Texas. "In essence, the need for security guard staffing is significantly reduced and replaced by technology and consolidation of technology."Smart video
The shift toward wireless IP-based surveillance systems has spawned more than cost reductions. The second major industry trend is the emergence of video analytics, or "intelligent video." This technology automates video monitoring with software that "watches" video feeds.
Essentially, computers attempt to detect certain behaviors or scenarios — abandoned objects or lurking persons — then automatically alert security staff.
Intelligent video is driven by increases in camera network complexity. As camera networks expand, it becomes increasingly difficult for security staff to effectively monitor all incoming video feeds. Intelligent video relieves the operator from the impossible chore of watching multiple monitors.
"Computers never lose attention, so video analytics remedies this problem," Chambers says.
Scribani says automatic software-based interactive alarms of event-based occurrences increase operator efficiency. Studies by Sandia National Labs have shown that security personnel are ineffective at video monitoring after only 15 minutes.
Another driver is the heightened state of security necessitating increases in security.
"Video analytics provide an effective first response mechanism, alerting law enforcement to possible problem behaviors in real time, possibly averting dangerous incidents," explains Chambers.
Using advanced software, these systems are able to pick up subtle changes in monitored areas, such as identifying someone leaving a package as small as a paperback book or, conversely, detect whether something had been removed.
However, system designers want more than just the ability to detect abandoned satchels. They want computers able to anticipate trouble by recognizing anomalies like suspicious behavior.
Researchers at the University of Maryland, among others, are pursuing this goal. Electrical and computer engineering professor Rama Chellappa is designing intelligent surveillance systems that include recognition of human gait, faces and behaviors.
Crunching data from surveillance cameras with sophisticated algorithms, Chellappa has developed a signature for characterizing human gait and corresponding activities, such as humans carrying objects like backpacks, handbags or briefcases.
When a person's limbs are unencumbered, gait movements are symmetrical. Represented graphically, these movements form a twisted helical pattern resembling a "figure 8" that Chellappa says is slightly different in each individual. He calls it "human gait DNA."
Since an individual's gait pattern is changed by any activity that changes the symmetry of the movements, for example carrying a package, by defining these signatures, the system may one day recognize unique patterns in human gait and automatically detect asymmetric movement, such as an individual walking with a hidden object tied to an ankle or wrist.Bandwidth bottleneck
One bottleneck to moving high-quality video data is limited network bandwidth. High-definition video requires high-performance networks to move massive amounts of data, which typically do not exist for this purpose.
Another Maryland researcher may have a solution. Electrical and computer engineering professor Chris Davis is working on advanced high-performance wireless links using both lasers and radio frequencies. Davis' network currently transmits television-quality video at more than 1 GB per second.
"Very few wireless links have these data rates," Davis explains. "We're moving high-quality video across campus at about 30 frames per second. Now, because the images are of such high quality, we can start exploiting them using algorithms to look for events."
Davis' system already can detect abandoned objects, read license plates at hundreds of meters, and identify types, sizes and colors of vehicles. Davis believes they're the only group that can grab high-definition video, move it around and process it in real-time.
"Most people are analyzing tapes after the fact to see if they can recognize anything," he says. "We can literally identify a vehicle of interest, alert police and stop it before it leaves campus."
Elaborate video networks are already emerging. In the United Kingdom, one network links 14,000 message signs, telephones and more than 6,000 surveillance cameras to traffic control centers. While primarily designed to provide a road safety monitoring system, it also gives law enforcement the ability to integrate video with a database of stolen car license plates.
"Now when a stolen car is located on the highway, local police are alerted and, using situational awareness technology, the system controls traffic lights along the pursuit route, giving police the opportunity to stop perpetrators safely without resorting to high-speed chases," says Richard Howes, CEO of Steelbox Networks in Atlanta, Georgia.Yet to come
While there has been a dramatic increase in the number of public video surveillance systems, Philip Blackmon, executive vice president of CompuDyne in Annapolis, Maryland, says the real surge is yet to come.
"There will be redundant systems, innumerable overlaps, and we'll be able to dissect a real-time occurrence from a variety of angles and with a variety of analytic considerations," Blackmon predicts.
"We'll have better quality, higher-resolution images. We'll be able to more effectively zoom in on details or pan out to take in macrocosmic concerns, and we'll have steadily improving analytical tools."
Blackmon believes deterrence values increase as technology evolves. The Dallas (Texas) Police Department knows this first hand.
In March, Dallas activated a wireless video surveillance network, an initiative designed to the help reduce crime in its central business district. Before Dallas implemented this solution, they set up a pilot in another area downtown called Deep Ellul.
"We noticed a perceptible drop in criminal activity," says Chief of Police Tom Lawrence.
The network now is deployed over about 30 percent of the downtown area, so Lawrence expects even greater deterrence.
"We expect to deter everything from the most serious felonies to misdemeanors," Lawrence says.
Broader adoption of next generation surveillance throughout law enforcement, however, is often hampered by cultural resistance to change among security directors.
"This is definitely true in the U.S. where many traditional security people balk at the learning curve involved in IP video," Hinkel says.
Lawrence, however, encourages police departments not to hesitate to go wireless.
"Wireless IP technology has come of age," he adds. "When we operated our pilot, we relied on wired technology. This was a mistake because your wired network is only as strong as its weakest link." The wireless system is self healing — if one link loses power, the video signal still goes through without any interruption in service.
To overcome techno-shock associated with advanced networking, Lawrence recommends fining a technology partner that can vet the various solutions available.
"Most police departments don't have the necessary expertise to explore and select the right technologies," he says.
Douglas Page ([email protected]) is a science and technology writer living in Pine Mountain, California.