摘要
车型识别是智能交通系统中的一个重要组成部分,近年来,车型识别技术已成为国内外研究热点之一。提出了一种基于特征车的车型识别方法。该方法首先对车辆图像进行预处理,然后通过K-L变换构造特征车,得到降维特征子空间,最后利用BP神经网络进行车型识别。
Vehicle recognition is an important part of intelligent transportation system. Now, the technology of vehicle recognition is becoming a hot topic all over the world. In this paper, we present a kind of method of vehicle recognition based on eigen-vehicle. Firstly, we preprocess the vehicle images. Then we build the eigen-vehicle through the use of K-L transform and gain a space with lower dimension. At last, we recognize the vehicles using neural network.
出处
《信息技术》
2004年第5期44-45,48,共3页
Information Technology
关键词
车型识别
K—L变换
神经网络
vehicle recognition
K-L transform
neural network
