Research On The Traffic Surveillance Technology With The Novel Magneto Sensor

The acquirement of the road traffic condition is the important precondition of traffic management, such as traffic flow control and guidance, as well as the necessary basis of the institution concerning traffic security measures, including traffic safety management, traffic accident detection and analysis of traffic accident causation. Therefore, the acquirement of the road traffic condition remains the fundamental significant issue traffic management, security, surveillance and maintenance of traffic infrastructure.This thesis relies on the Item of High—tech R&D Program (863)—"The New-type Sensor concerning Traffic Condition Acquirement; Optimize and Amalgamation of Sensor's Network", chooses the traffic condition acquirement technology of the new-type sensor as the research object, and focuses on the detection of traffic flow, time occupancy ratio and vehicle instantaneous speed, and technology of vehicle model identification. The thesis is constituted of the following related parts:The paper makes use of the digital filtration technology to handle the noises removal of data collected by the sensor, and gets the corresponding traffic condition through the algorithms of traffic flow detection and time occupancy ratio detection. During the research of the traffic condition acquirement, this paper brings forward a method of instantaneous speed detection which based on the new-type double modules sensor. This method takes the instantaneous speed as a direct acquire quantum, and carries through the experimental verification. During the research of vehicle model identification, the thesis puts forward a method concerning vehicle model classification, which takes the vehicle wheelbase as normal standard. This method analyses category of the vehicles through the calculation of axle's length, and also carries through the experimental verification. Moreover, this thesis puts forward a method of vehicle model classification, which uses the characteristic of vehicle models as standard. This method makes use of the Decision Tree to classify the vehicles through the character analysis of data collected by the sensor, and carries through the experimental verification.