Predictive crime analysis DigiSensory cameras predict crime before it happens

Published 25 March 2011

With DigiSensory Technologies’ sophisticated cameras and sensor systems, law enforcement agencies and transportation departments across the United States are now able to proactively monitor and respond to crimes or accidents as they unfold; the company’s Avista sensors process the images that its 3.2 megapixel high-resolution camera records in real time and can automatically detect when a crime is occurring; when it senses something it will alert law enforcement officials instantly; the sensors can also assist transportation departments in analyzing traffic patterns in real time; the system could allow officials to change one way streets, design real time traffic signals, and multiple speed limits to make traffic flow more smoothly

With DigiSensory Technologies’ sophisticated cameras and sensor systems, law enforcement agencies and transportation departments across the United States are now able to proactively monitor and respond to crimes or accidents as they unfold.

In 2009 the East Orange, New Jersey police department was the first U.S. agency to install DigiSensory’s Avista Smart Imaging Sensor system as part of its broader network of security cameras around the city.

The Avista sensors process the images that its 3.2 megapixel high-resolution camera records in real time and can automatically detect when a crime is occurring. When it senses something it will alert law enforcement officials instantly.

Jose Cordero, the East Orange police director, says that the cameras have helped reduce response times from minutes to seconds.

“It’s no different from what we do on the street, but now we have a system in place that can look at our data and turn information into intelligence in real time,” Cordero said.

This is particularly critical as East Orange’s streets are riddled with gang and drug related violence.

When Cordero joined the force in 2004, he says East Orange had a crime rate fourteen times the national average. Since then, however, violent crime has dropped by two thirds, largely as a result of the department’s $1.4 million camera system.

In 2009 East Orange installed ten Avista Smart Cameras. These cameras have sensors that can identify behavioral patterns and actually predict when a crime is going to occur. This information is then stored in a database tied to geographic location so that it will be able to forecast where a crime is most likely to occur next.

“The system will predict when the next likely event will occur at these locations during these particular times during this particular day,” Cordero says.

According to their Web site, Avista sensors are “programmed to analyze the environment, recognize specific user defined patterns or profiles, classify profile as a threat category and warn or alert of common and specific elements of street crime and help solve crime by providing video footage of completed or potential crimes.”

Craig Primiani, vice president of sales of DigiSensory’s North America division, says the technology takes a “proactive approach to law enforcement, so that officers are alerted as an event occurs, rather than after.”

He explains that unlike existing video systems which simply record events and store them in an archive to be reviewed only after a crime occurs, Avista sensors analyze events as they occur.

Primiani says that DigiSensory’s technology is applicable beyond just law enforcement and can greatly assist transportation departments in analyzing traffic patterns.

DigiSensory is in the final stages of installing a traffic camera and sensor system for Washington D.C.’s Department of Transportation (DCDOT). The system will provide DCDOT with real time analysis of traffic patterns to predict commute times.

Primiani believes that the system has vast future potential.

“There’s so much information we can gather about traffic patterns. If D.C. had 1,000 of these they could do predictive analysis. They could make decisions on one way streets that change from one way in the morning to the other way at night, design real time traffic signals, and multiple speed limits,” he said.

In short, smart sensors can provide cities with “information to make real time traffic decision to make the city’s existing roads more effective.”

In addition, these traffic systems could be integrated with existing law enforcement security cameras and vice versa. So if someone is hit in an accident, the cameras could automatically alert emergency responders.

To allay critics who charge that the camera systems are too invasive or violate privacy laws, the company has built in privacy safeguards.

Primiani explains that, “We have the ability with a flick of a switch to black out the objects we are tracking.”

The sensors can outline an object and make its specific characteristics indistinguishable.

“For instance, if a car is going down the road, with existing systems we can see that it is a 1984 Chevy Nova, but we can block it so that we do not know what type of car it is. We’ll just know it’s a car and the same thing applies to humans,” he said.

p

David Young, a spokesman for the company, says that the technology has a broad appeal and can be used in many different situations.

“The technology has wide application in law enforcement, school safety, government institutional safety, crowd management, transportation departments, prisons and more. With bullying being one of the biggest problems occurring in American schools today, imagine if principals would be automatically notified of a potential incident moments after students began to loiter,” Young said.

Government agencies and business around the world have already taken notice of DigiSensory Avista predictive sensors.

Louis Vuitton stores in China have already installed these cameras, while the U.K. Olympic Organizing committee is considering implementing these systems for the London 2012 games.