University of Florida researchers believe they have outwitted the makers of radar detectors and other speed detection countermeasures. Using a $5 million grant from the Air Force, they've developed software that can use live video camera footage to recognize objects and measure their speed. It does so by calculating the distance a car moves in a series of photos compared to a fixed object -- such as a light pole -- in the background. Because the system is entirely passive, unlike conventional laser and radar speed guns, it cannot be detected.

"If it can view the object moving, that's all it needs. The computer figures out everything else," said Warren Dixon, University of Florida assistant professor of mechanical and aerospace engineering. "We're trying to use both regular and infrared cameras, so night or adverse weather conditions don't present a problem."

The system is similar to the SPECS speed averaging cameras in use in the UK and Australia. The latter system uses multiple cameras to calculate a car's speed over a known, fixed distance. The researchers believe they can create a device has two big advantages over SPECS -- it would be entirely mobile and it would work with a single camera.

The military continues to be interested in potential applications for the software. Additional details on their research will be published in the latest issue of IEEE's Transactions on Robotics and Automation.