基于SVM的车型识别系统的设计与实现

Design and Implementation of Car Model Identification System Based on Support Vector Machine

支持向量机(SVM)理论建立在结构风险最小化原理基础上,对非线性、高维数的小样本问题有非常好的分类效果和学习推广能力。本文设计了基于支持向量机的车型识别系统,系统通过对摄像机采集的视频图像进行运动目标检测分割、特征提取与选择、模式识别等处理,达到实时车型识别。试验结果表明,该系统有很高的识别率和适应性。

Support vector machine （SVM） has good ability of classification and generalization for the nonlinearity, multi-dimension, and small-sample problems. The paper presented a Car Model Identification System （CMIS） based on support vector machine which manipulated the video image acquired from video cameras by employing moving object detection algorithm, feature extraction and pattern recognition algorithm. Experiment results show that the system has very high identification performance.