'Smart' Cameras Interpret Behavior

Ohio State University researchers are developing a “smart” surveillance system that will be able to determine if a person on the street appears to be lost or is acting suspiciously, according to the student daily newspaper, The Lantern. The goal is to create a network of smart video cameras that will allow officers to quickly and efficiently observe and monitor a wide area.

“In my lab, we’ve always tried to develop technologies that would improve officers’ situational awareness, and now we want to give that same kind of awareness to computers,” said James W. Davis, OSU professor. Davis said the goal is to analyze and model the behavior patterns of people and vehicles moving through a scene. “We are trying to automatically learn what typical activity patterns exist in the monitored area and then have the system look for atypical patterns that may signal a person of interest.”

The system will focus on where a person goes and what they do. The first algorithm expands the small field of view that traditional pan/tilt/zoom cameras provide by taking a series of snapshots from every direction within a camera’s field of view and combining them into a 360-degree, high-resolution panorama.

The operator can click anywhere on the picture and the camera will pan and tilt to that location for a live image. Another program will map locations onto an aerial map of the scene and then calculate where the view spaces of the security cameras overlap and determine the geo-referenced coordinates of each ground pixel in the panoramic image. A third program will use the aerial and panorama views for tracking people, which could be used to instruct a camera to follow specific people based on their behaviors.

Stay Informed Join our Newsletter

Having trouble finding time to sit down with the latest issue of
ELECTRICAL CONTRACTOR? Don't worry, we'll come to you.