Research Of Traffic Flow Detection Method Based On Geomagnetic Sensors

With the increasing of vehicle and the development of the transportation management, the detection of the vehicle information becomes more and more important in Intelligent Transportation System. At present, the main devices for vehicle detection are using inductive loops in our country. Single detector in vehicle distinguish is unable to accurately detect vehicle speed, the method can only be used statistical models, and which is also an important reason leading to errors. Using geomagnetic detector is a new type of vehicle information detection based on the changes of magnetic field. Under the preprocessing and character analysis of the collection data, this dissertation proposes the method of traffic amount detection based on geomagnetic detections. Due to various factors and environmental influences in practical application, the collection data are caused a large degree of deviation, and thus impacts the data processing and road information judgment. Wavelet transformation has self-adaptability to the signal which shows lower time resolution and higher frequency resolution in low-frequency part, and lower frequency resolution and higher time resolution in the high-frequency part. Therefore because of the good characteristics of the wavelet transformation, the data can be pretreated. In this dissertation, author uses the method of traffic amount detection through the threshold value and vehicle statistics and distinguishes based on magnetometers detections and give an online experiment result which shows highly accuracy. The detection system is a well-organized complex system on the system designing. The operation of the detection system is constitutes of different modules such as data acquisition,preprocessing and analysis, road information identification and transmission. At the end of the dissertation, the overall system structure is presented and the communication protocol and demonstration program are established.