An image processing system to measure vehicular queues and an adaptive traffic signal control by using the information of the queues

We have developed an image processing system to measure vehicular queues in approach lanes of an intersection. TV cameras of this system are installed at a height of 10 m or so at an intersection to take images of traffic inflows at a bird's-eye view. This system using a macroscopic detection method allows measurement of queue lengths up to 300 m long if the visibility is good. We also propose a traffic signal control method to adjust signal split using the information of vehicular queues at a critical intersection. We have done simulation experiments to estimate the control efficiencies. The results show that the proposed method follows the optimal split even if the traffic inflow patterns change, and prevents increasing the delay.