Law enforcement technologyIdentifying bad guys in a crowd
New CCTV technology aims to do two things: for post-incident forensic analysis the system retraces a person’s steps after they have left a site; and for real-time analysis the system alerts security to immediate or imminent threats; researchers say the technology will also address privacy concerns: if a computer could automatically detect and preserve footage “of interest,” the images of the rest of the people captured by CCTVs can be safely deleted, minimizing the invasion to privacy
In early June the Engineer reported that security company Ipsotek had performed a trial of its new Tag and Track system at Manchester Airport with support from BAE Systems and Kingston University.
The system aims to do two things: for post-incident forensic analysis the system retraces a person’s steps after they have left a site; and for real-time analysis the system alerts security to immediate or imminent threats.
Dr. Boghos Boghossian, technical director at Ipsotek, told the Engineer that as well as tracing people retrospectively, the system can work in real time by tagging and tracking people after passport scanning – if the cameras are located at an air port — or when they exhibit certain suspicious behaviors.
Two weeks ago the Engineer reported that researchers are devising ways automatically to analyze CCTV and other security footage. They hope that the technology will help police and other security agencies catch bad guys more quickly and more often, while minimizing the invasion of privacy of law-abiding citizens.
Researchers from Kingston University’s Digital Imaging Research Center (DIRC) are working on the technology. James Orwell, head of the DIRC’s Surveillance Research Group, told the Engineer, “We plan to develop components to automatically analyze multi-camera networks and footage before and after a trigger incident, such as a riot or fight, to produce a set of video segments relevant to a potential police investigation.” Using visual analysis (the DIRC offers few details on the exact mechanism), the software can scan all the relevant footage and largely automate a process that police might spend countless hours doing by hand.
The researchers note that not only does the technology make cops’ task easier, it also helps address some of the concerns of privacy advocates, who are uncomfortable with the amount of video footage of people’s behavior that winds up in archives. If a computer could automatically detect and preserve footage “of interest,” the images of the rest of the people captured by CCTVs can be safely deleted, minimizing the invasion to privacy.
