Technology is working for the British police these days. Last week I wrote about a forensic lab in London that has developed an innovative method for guaranteeing that audio recordings are genuine and not doctored. Now here’s new research out of the University of London that may solve a tricky reconnaissance problem: how to track suspects as they move from one security camera to the next.
In the industry this is known as the problem of tracking people across “non-overlapping” camera networks, or the fact that often there are blind spots between camera views. Most solutions rely at least partially on appearance matching—using distinguishing visual features to track a suspect as he moves between security cameras. But this approach doesn’t work in instances where the quality of a feed isn’t good enough to pick up details.
So a pair of engineers, Riccardo Mazzon and Andrea Cavallaro, have come up with a new solution. They’ve devised an algorithm that predicts the paths people are likely to take in the blind spots between security cameras, and the places they’re likely to reappear once they’re back on tape. The algorithm takes into account layout features that influence walking patterns, like exits, seats, and meeting points, and also behavioral data about pace and the routes people tend to take in different kinds of walking environments.
They tested their model using security feeds at the arrivals terminal at London Gatwick airport. In different tests they were able to correctly re-identify people from one security camera to the next about 50 percent of time—not bad for a first run, and all without any appearance matching. And if you’re wondering what you can do to confound this system, here are two tips: The algorithm was particularly ineffective at re-identifying people who exited a camera frame at the same time, location, and pace as another person, and it tended to lose people who started to walk dramatically faster or slower between camera views.
You can read the whole study, complete with nifty motion-tracking graphics, here.
Image courtesy of Soctech.
The author is solely responsible for the content.
Leon Neyfakh is the staff writer for Ideas. Amanda Katz is the deputy Ideas editor. Stephen Heuser is the Ideas editor.
Guest blogger Simon Waxman is Managing Editor of Boston Review and has written for WBUR, Alternet, McSweeney's, Jacobin, and others.
Guest blogger Elizabeth Manus is a writer living in New York City. She has been a book review editor at the Boston Phoenix, and a columnist for The New York Observer and Metro.
Guest blogger Sarah Laskow is a freelance writer and editor in New York City. She edits Smithsonian's SmartNews blog and has contributed to Salon, Good, The American Prospect, Bloomberg News, and other publications.
Guest blogger Joshua Glenn is a Boston-based writer, publisher, and freelance semiotician. He was the original Brainiac blogger, and is currently editor of the blog HiLobrow, publisher of a series of Radium Age science fiction novels, and co-author/co-editor of several books, including the story collection "Significant Objects" and the kids' field guide to life "Unbored."
Guest blogger Ruth Graham is a freelance journalist in New Hampshire, and a frequent Ideas contributor. She is a former features editor for the New York Sun, and has written for publications including Slate and the Wall Street Journal.
Joshua Rothman is a graduate student and Teaching Fellow in the Harvard English department, and an Instructor in Public Policy at the Harvard Kennedy School of Government. He teaches novels and political writing.