Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now



San Francisco Subway to Get ‘Behavior Recognition’ Cameras

The software builds memories to determine behavior that is reportable

San Francisco’s Municipal Transit Authority, which oversees the city’s railway trains, has contracted BRS Labs, a company that produces behavior recognition software for video surveillance, to install cameras containing that software at 12 subway stations when the Central Subway project is complete in 2019.

Construction on the Central Subway project began in 2010 and the transit authority has contracted for up to 22 cameras, linked in a video management network, to be installed at each subway station—totaling 288 cameras—that can be monitored by transit authority staff at each station, BRS Labs says.

The behavior recognition software will be incorporated in each of the cameras at the stations and the video from each camera will be streamed to the “Behavior Recognition System” where it will be analyzed, BRS Labs says. The analysis will focus on “behavioral patterns, activities and content in the scene” and report “unusual or abnormal behavior,” the firm says.

How the behavior recognition software determines behavior that is reportable is the video systems build memories of observed behavior patterns that mature with time, BRS Labs says. The system “has the capability to learn from what it observes, remember activity patterns and adjust to changes in the environment, field of view and equipment without manual [human] interaction.”

When reportable behaviors or situations—such as abandoned bags—are observed, the system will send alerts to “Train Control” and station personnel, but also to mobile devices such as smartphones and tablets, BRS Labs says.