Purdue VACCINE Working with Police on Hazmat Tools
March 22, 2012
Researchers are creating applications that enable cell phones and other portable devices to translate the meaning of gang graffiti for law enforcement and hazardous materials placards for emergency first responders.
The researchers presented their prototypes on Thursday (Aug. 18) to law enforcement and homeland security officials in Indianapolis.
"The whole idea of these projects is that you literally take a picture of the gang graffiti or a hazardous materials placard with a mobile phone and then the system interprets the images," said Edward J. Delp, Purdue University's Charles William Harrison Distinguished Professor of Electrical and Computer Engineering.
Chief Gary Coons at the Indianapolis Department of Public Safety Division of Homeland Security said the mission is to determine and develop the future technology needs of law enforcement.
The projects are being introduced as GARI, for Gang Graffiti Automatic Recognition and Interpretation, and MERGE, for Mobile Emergency Response Guidebook.
This work is funded by the U.S. Department of Homeland Security Science & Technology Directorate (S&T) and is part of the Visual Analytics for Command, Control and Interoperability Environments Center of Excellence, or VACCINE, headquartered at Purdue.
"Having university researchers in the centers work directly with potential user organizations allows S&T to identify real problems and then to move technological solutions into the field quickly," said Joseph Kielman, DHS program manager for VACCINE. "Nurturing such partnerships means that the research is already primed to respond as needs arise, rather than being developed anew."
VACCINE and Purdue are collaborating with the Department of Public Safety Division of Homeland Security to bring this technology to the Indianapolis Metropolitan Police Department to interpret gang graffiti. Over the last eight months VACCINE researchers have worked with IMPD officers on the gang-graffiti system, said Timothy F. Collins, managing director of VACCINE.
"Gang graffiti basically tells a story," Collins said. "Investigators want to not only catch who put it there but also to understand its meaning. Sometimes they indicate when someone is about to get killed or whether a rival gang has moved in that could lead to an increase in crime. An officer might take a picture of graffiti and ask the system to show all the similar graffiti that has occurred within 2 miles of the location."
The system uses image analysis algorithms to analyze the graffiti, while also identifying the GPS coordinates.
"You take a picture with an Android mobile phone, and it not only records the image, it also records the GPS coordinates, the date and time," said David S. Ebert, Silicon Valley Professor of Electrical and Computer Engineering and VACCINE director. "It can provide some analysis right on the phone and also can access a more extensive database of graffiti on a server."
Law enforcement and public safety officials will test the portable systems in the field.
"Interpreting these images is not a trivial task," said Delp, who is working with Mireille (Mimi) Boutin, an associate professor in Purdue's School of Electrical and Computer Engineering, and two graduate students.
The image analysis procedure includes evaluation of features such as colors and shapes of characters in gang-graffiti images. The information is sent to a server and compared against the graffiti image database. The matched results are sent back to the device where the user can then review the results and provide extra details to refine the analysis.
"Once the graffiti is completely decoded and interpreted, it is labeled and added to the database," said Collins, a former Indiana state trooper. "Gangs are a serious threat to public safety throughout the United States. Gang members are continuously migrating from urban to suburban areas and even rural areas. They are responsible for an increasing percentage of crime and violence in many communities."
Street gang graffiti is a common way to communicate messages, including challenges, warnings or intimidation to rival gangs.
"It is, however, an excellent way to track gang affiliation and growth, or even sometimes to obtain membership information," Collins said.
The Department of Public Safety Division of Homeland Security and the Indianapolis Metropolitan Police Department will be the first in the United States with this technology.
In the MERGE project, the system automatically interprets the hazmat signs and also draws a recommended evacuation zone, depending on the dangers posed by the materials. The signs and placards are displayed on vehicles and structures containing hazardous materials.
Responders now use an Emergency Response Guidebook to identify the placards.
"This guide assists those in an emergency with the knowledge of how to handle hazardous materials," Collins said. "However, as one might expect, the guidebook is large and requires precious time to search an index to determine the best way to handle a particular hazardous material."
The MERGE system is an electronic version of the guide and includes new features and capabilities.
"These new capabilities include the use of image analysis methods to automatically determine the type of hazardous materials present based on an image taken of the sign or placard, as well as the appropriate response protocol and evacuation perimeters," Ebert said.
The researchers developed "lightweight algorithms" that operate quickly, have low-energy consumption and require low memory.
VACCINE plans to expand deployment to other interested agencies this fall.
Writer: Emil Venere, (765) 494-4709, email@example.com
Sources: Edward J. Delp, 765-494-1740, firstname.lastname@example.org
Timothy F. Collins, 765-494-0536, email@example.com
Shelly Higdon, 317-985-2324, Indianapolis police, Shelly.firstname.lastname@example.org,
Elia James, Division of Homeland Security Department of Public Safety, 317-506-7770,
David Ebert, 765-494-9064, email@example.com
Mireille Boutin, 765-494-3538, firstname.lastname@example.org