The Design And Implementation Of Tram Anti-collision Method’s Key Technologies

As the trams share road right with social vehicles on urban roads, pedestrians and vehicles may cross the track, causing collision accidents. The existing protection is mainly for highispeed trains and metro trains. Therefore, a tram anti-collision method based on image processing is discussed in this paper. The system sends anti-collision warning message to the driver through the identification and distance measurement of the social vehicle in front of the tram. It can avoid the occurrence of traffic accidents and ensure the safety of travel.In this paper, the anti-collision warning method is designed by using sensor technology,communication technology, image processing method, machine learning theory and embedded technology. It has two parts, which named image processing controller and ranging-communication coprocessor. The image processing part is used to receive the information from the image sensor, which uses ordinary camera. This part identifies the social vehicles through scale-invariant feature transform and support vector machine method. This method process images of samples through build image pyramid, locate key points, assign main direction of vectors and generate them. The words of bag model is built with clustering method. The words of bag model is used as the input of the support vector machine to generate the vehicle identification classifier. With the classifier, the system can identify social vehicles and send signal to the coprocessor. The coprocessor uses the ARM chip with the Cortex-M3 architecture and designs the peripheral circuit. The coprocessor connects with on-board computer by configuring the wireless communication network. When receiving the ranging signal, the coprocessor calculates the distance of the social vehicle and sends it to the on-board computer.The driver receives warning signal from the on-board computer. The anti-collision system is tested in a laboratory environment. Experiments show that the system can identify of the vehicle in front of the tram, measure the distance and send the information to the on-board computer. When a collision accident may occur, the on-board computer can send a speed limit or brake command to the tram, which achieves the design requirements.This system achieves anti-collision function of tram based on image processing method and support vector machine. The system has the characteristics of simple structure, low cost,which provides a certain reference for the development of tram protection system.