Tel Aviv University researchers have teamed up with General Motors to invent a new hi-tech system to help keep drivers safer on the roads, and they are working on upgrades to fine-tune the technology even more. Professor Shai Avidan, a member of the university's Faculty of Engineering, is an expert in image processing who is working with researchers at General Motors Research Israel on a new rapid alert system to prevent accidents on the road. According to the Israeli traffic safety advocacy group Or Yarok (Green Light), there were approximately 33,000 traffic accidents on urban roads from 2007-2009. Fully one-third – 11,000 -- occurred between 1:00 p.m. and 6:00 p.m., hours when children were returning home from school or going to other activities, and adults were traveling home from work. In the same period, a total of 97,500 people were injured in traffic accidents. Avidan and his team have developed software for cameras to be mounted on GM cars that will detect moving threats on the road, allowing a driver to make an instant decision about how to handle the risk. The technology works on infrared, grey scale and color cameras. The research, published in the IEEE Transaction on Pattern Analysis and Machine Intelligence, has also been presented at various professional conferences. The challenge still faced by the research team is that of finding a way to enable the camera to differentiate between the movement of human beings, animals, and inanimate objects such as other vehicles on the road. Avidan is hoping to use research he previously participated in, that led to the development of the now-commercial MobilEye smart camera system – to create even “smarter” cameras. The MobilEye system detects vehicles and tracks them in real-time, but Avidan would like to develop a camera that can also categorize the moving objects it detects. “Such a tool could double check for vehicles in your blind spot, help you swerve when a child runs into the street, or automatically block your door from opening if a cyclist is racing toward you,” he explains. A more futuristic approach predicts the day when the same technology might be used in surveillance to detect a potential intruder, or in computer games to track a player's movements.
