Study On Decision-Making Algorithm Of Vehicle-Mounted Video Variably Transmission Based On Information Fusion

With the development of social economy and technology, highway transportation make our lives convenient, but their security risks have been increasingly exposed at the same time. Highway safety accidents frequently occur and bring the great harm. As an information carrier, video has been the main methods of monitoring transportation. However, the public mobile network’s expenses are relatively high. In this paper, the algorithm about the carried the vehicle mounted video variably based on information fusion was studied, and the contents were organized as follows:(1) The characters of information fusion and Back-Propagation (BP) neural network are made clear, and the DS-BP decision algorithm is proposed, which combines with Dempster-Shafer theory (DS) and BP neural network for BP neural network’s fixed structure couldn’t acclimatize to complex environment.(2) Combined with the reasons of vehicle’s abnormal operation, extracting the feature of the vehicle’s video, audio, speed, acceleration, steering strength, coolant temperature, engine speed and so on, which lays the foundation for future vehicle-mounted video’s variable transmission and decision-making.(3) The initial of BP Neural Network was optimized by Particle Swarm Optimization (PSO), to improve the problem of BP Neural Network trapping into the local minimum and converging very slowly.