Traffic Control in the Smart City Using Video Cameras and Analytics
The traffic light was invented in 1868 and the first one was installed in London, outside the Houses of Parliament. It was operated by gas and, sadly, blew up, killing a policeman. According to the United States Access Board, there are more than 300,000 traffic lights in the US today and a good estimate worldwide would be in the millions.
Most traffic signals today are controlled by counting the number of cars that go over a wire induction loop buried in the street. The metal in the car disturbs a magnetic field and a counter is incremented, unless you’re riding a motorcycle that is. The induction loops often aren’t sensitive enough to detect your bike, and you’re left standing by the light, keeping the engine ticking over. Induction loops are also difficult and expensive to maintain or repair. You have dig up the road to replace them when they go wrong.
Step forward the video camera. By mounting video cameras on light poles above the street, equipped with video analytics software to count the vehicles passing by, traffic signals can be easily and efficiently controlled. Video cameras are much cheaper than the induction loops and easy to adjust or repair. They can transmit their video stream to a small server in the roadside cabinet and the analytics can then be sent to the controller in the same cabinet. It’s easy to program for multiple lanes, turns and rights-of-way, and by adding analytics which can recognize public transit or emergency vehicles, these can be given priority too.
Although video camera-based systems can be expensive, they can relieve the need for other even more expensive measures like new roads. One of the other advantages of video cameras is that they can also be used for incident monitoring.
