Abstract
Among the most impotant research in Intelligent Transportation Systems (ITS) is the development of systems that automatically monitor traffic flow at intersections. Such systems would be useful both in reducing the workload of human operators and in warning driver and pedestrian of dangerous situations. The might be able to reduce the number of accidents and protect safe of driver and pedestrian. In our system, we employ algorithm of moving objects segmentation to extract moving objects information that give a new object ID to those object piexls. Each ID have own features, such as position, area and motion vector. Then we track each ID and employ least-square solution and linear or quadratics equation to request future trajectory. We can distinguish that moving objects are car or motorcycle or pedestrian and understand situations of traffic. When a car to make a turn and pedestrian to pass through intersection can predict whether they might lead to accident.