| Since the reform and opening policy in past decades, great changes have taken place in economy, especially the urban construction. Urban transport plays an important role in the development of cities. Today, with the increasing amount of city car, the hurricane development of city car velocity leads to a series of problems hindering the urban development. It includes traffic congestion, lack of energy, environment pollution, accidents frequently intensified and so on. How to improve the utilization of urban traffic data becomes necessary to solve the above-mentioned aspects of urban traffic problems. In view of the urban public transport and private vehicles compared with passenger capacity, it requires relatively small investment, small footprint; it operates with high operational efficiency, less pollution. the government should spare no effort to develop the urban public traffic, and achieve digital and intelligent goals. So the only way to improve the urban traffic are to improve the efficiency of public traffic operational management and social service level.This paper aims to obtain bus passenger traffic data based on multi- sensor array pedal. BP neural network algorithm is used for original data preprocessing and intelligent recognition, calculating on the number of passengers getting on and getting off the bus.This paper analyzes the footprint variation of passengers getting on and off the bus. It obtained characteristics of sensor data. Traffic data criterion is proposed based on foot contour and gather the data processed by the neural network method by human motion movement principles for direction recognition. Ultimately identify the number of passengers to get on and off. The hardware design implementation and identification software algorithm applied in this paper was elaborated in detail. Finally, a systematic test of the algorithm was operated. Test result shows that the accuracy rate is 93% which indicates the system is with strong reliability. |
PDF Full Text Request