Automotive vehicles are an integral part of modern society in the US and around the world. As the number of vehicles on the road increases, challenges like traffic monitoring, analysis, safety, and security become increasingly relevant to municipalities, roadway engineers, law enforcement, and others. New sensing and control systems are needed to achieve improvements in these areas.
In the past five years it has been increasingly clear that expensive, invasive in-ground sensing can be replaced by low-cost, high coverage video technology. By aiming a surveillance camera at a freeway, roadway, intersection, or other vehicle path, one can capture video of traffic density, vehicle behavior, appearance, and more. This provides much more data than simple road-based sensors. When combined with intelligent algorithms and computation resources, this video data is a very rich source of information!
Quantum Signal has developed algorithm technology and software that detect, tracks, extracts, and analyzes motor vehicles captured on video. Initially developed as an off-shoot of our advanced biometrics program, the Quantum Signal Vehicle Tracking Framework (QSVTF) allows sophisticated tracking of automotive vehicles - cars, trucks, buses, tanks - using any surveillance camera. The system will track and monitor, generate statistics on path length/speed, count numbers of vehicles passing through an area, and provide a multitude of other metrics. QSVTF is robust to the angle of the cameras, and does not require a complex installation or the mapping of road geometry (as some other systems do), and is compatible both with day and night operation. QSVTF also allows completely flexible monitoring of specific areas within a video so a lane of interest or specific turning area might be monitored while ignoring other areas. Graphical definition tools have been developed to enable even a novice to specify the analysis area. This is not simple motion detection - this is real image processing that comprehends a wide variety of traffic monitoring situations and control.
QS has also developed advanced technology that recognizes the make and model of vehicle. This technology is not based upon the more common automatic license plate recognition (LPR); rather, it uses the appearance of the vehicle in a still or video to analyze type. The developed algorithms work in concert with the QSVTF described above, and can be used at different angles under a variety of outdoor lighting.
To learn more about these capabilities and how they can be used to meet your needs or facilitate new capabilities, please contact us today!