关键词: 智能交通, 车辆识别, 导向无环图支持向量机, 各向异性磁阻

Abstract:

With the increasing number of motor vehicle, urban transport is widely assisted by intelligent transportation system (ITS),. Traditional vehicle detector cannot meet the requirements of ITS’s reflection on traffic condition, the function of vehicle classification has received great attentions. Current AMR vehicle classification detector classify vehicle by the length of vehicle but the vehicle structure. To solve this problem, this paper designs a new kind vehicle classification detector. A single detector can collect three directions’geomagnetic variation, and magnetic field strength changes in three-dimensional space can be gained with the combination of a high-precision time module, which enriches the collected information. By comparing the pros and cons of common vehicle classification algorithms, DAG-SVM is selected to classify vehicles. The experimental test results prove the effectiveness of vehicle classifier detector designed in this paper.

Key words: intelligent transportation, vehicle classification, DAG-SVM, AMR