KINGSTON, R.I. (January 26, 2004) -- Since most criminals only strike when they aren’t being watched, reliable surveillance of homes and businesses is a round-the-clock job. A University of Rhode Island researcher has made that job considerably easier and less expensive, thanks to a new technology he developed that can automatically track moving objects in real time.
Using low-cost, commercially available hardware, the Automatic Image Motion Seeking (AIMS) camera follows a moving object and keeps the target at the center of the field of view.
"This camera has broad impact for security surveillance, because it eliminates the need to have a full-time guard watching a video screen," said Ying Sun, URI professor of electrical engineering who began developing the device in 2002. "It’s one intelligence level above any other existing system, and we’ve found the right compromise between speed and accuracy."
It’s also inexpensive. Sun, a Wakefield resident, said that the system can operate on a $30 webcam as well as on more sophisticated equipment. It just requires a motor-driven, pan-tilt camera mount and a processor. Using low-cost equipment, the system could cost less than $300, making it ideal for many home uses. And because it can track movements, one AIMS camera can be just as effective as several stationary cameras.
At a rate of 15 frames per second, the camera analyzes images for any motion. Once a moving object is found, it feeds that information to the camera mount to begin tracking the object as it moves.
"We’re working on adding ‘behavior modifiers’ to the system as well, so that once the camera identifies motion it can be programmed to continue to track a given size, shape or color regardless of any other motion," Sun said. He also believes that a camera that can quickly track motion has a psychological effect on criminals. "If they see that the camera is following their movements, they may think that a security guard is manually operating the camera and is aware of their presence. It’s likely that the criminal would then decide to go elsewhere."
In addition to property surveillance at such places as ATM machines, businesses, warehouses, factories, and homes, the camera has applications for homeland defense, military uses, child monitoring, playground surveillance, border patrol, and video conferencing, among others.
"Existing video conferencing equipment requires the speaker to remain in one place in front of a stationary camera. With the AIMS camera people can walk around and the camera will automatically follow them," Sun said.
The technology is based on an image-processing algorithm for real-time tracking. Because of the effectiveness and computational efficiency of the algorithm, the feedback control loop can quickly achieve reliable tracking performance. The algorithm is implemented in the Visual C++ language for the Windows Operating System on a PC, however it could be configured to operate on an embedded PC, handheld computer or digital signal processor chip. Video recording can be triggered by the presence of motions and stored on a computer hard disk as AVI files. Motions can also trigger an alarm or other security measures.
Former URI graduate students Xu Han and Yu Guo worked with Sun on the project. All three are co-inventors of the AIMS tracking algorithm, which has a U.S. and international patent pending.
Cite This Page: