Data Analysis Techniques Supporting Intelligent Transportation System And Its Application Development

With the rapid development of economy and continuing progress of civilization,together with sensor technology,communication technology and computation technology,Intelligent Transportation technology is coming into being.It effectively applies and integrates advanced science and technology to the whole ground transportation management system,which in turn strengthens the ties between vehicles,roads and road users.The characteristics of all-round and all-weather makes it an integrated transportation management system which guarantees the safety,boosts the efficiency,improves the environment and saves the energy.Data collection plays a key role in ITS,which embodies how to gain and process raw traffic data.Two methods,static and dynamic traffic information collection,comprise real-time traffic information collection.ALPR belongs to static traffic information collection,and its fast identification of vehicles makes it widely used.However,due to illumination,climate,vehicle speed,obstruction and sensor dysfunctionality,its accuracy often deviated.As main means of dynamic traffic information collection,GPS technology saw its wide application home and abroad.Real-time three-dimensional coordinates,velocity etc.al can be provided.Currently its dominant users are buses and taxies.Implied in a massive amount of GPS traces,are the objective attributes and regular pattern of the city.Also,subjective driving biases are also covered in.In view of the above challenge,we propose a method for ALPR correction algorithm which works independent of the underlying recognition algorithm.Experiments show it can discover falsely recognized characters and classify the patterns of recognition error.Meanwhile,a real-time taxi anomaly detection method is proposed and implemented on our environment.Simulations show that it can effectively recommend appropriate routes and detect the anomaly in real-time.