The real-time perception of the road network operation in a low-cost way is the key factor to gurantee the transportation safety and the unblocked operation. The rapid and accurate detection of traffic incidents and the collection of traffic data are often difficult. The traditional method is not visual, real-time and cost-effective. In recent years, the development of video big data and artificial intelligence provides a new way to solve these problems. This paper reviews the research history and application limitations of the intelligent video analysis in the field of transportation, as well as the characteristic connotation of video big data, the design of an artificial intelligence platform driven by video big data, and the solution and the application for highway traffic.

traffic video big data  /  artificial intelligence  /  deep learning  /  cloud computing  /  intelligent highway