摘要： 根据模式识别理论和支持向量机(SVM)网络技术，对运动车辆的检测和模式识别、分类进行了研究，提出了基于双帧差“或”运算检测法和基于SVM网络的车型识别和分类算法。实验结果表明，所设计的检测方法和SVM模式识别方法能够快速有效地识别车辆类型并正确地进行分类。

关键词: 支持向量机, 模式识别, 车型检测, 车型分类

Abstract: Using pattern recognition theory and support vector machine(SVM) network technology, vehicles automatic detection and classification method is discussed in detail. An algorithm of vehicles recognition on the combination of double inter-frame difference ‘or’ operation and SVM network is presented. The experimental results indicate that the algorithm can recognize and classify vehicles rapidly and effectively.

Key words: Support vector machine(SVM), Pattern recognition, Vehicles detection, Vehicles classification