摘要
本文设计并实现了一个基于视频图象的车辆自动分类系统,从系统的构成开始,详细阐述了各部分的实现方法。系统中综合应用了运动边缘检测、图像二值化、闭包算法等方法对图像进行预处理,然后对车辆轮廓进行傅立叶级数展开提取特征,最后结合基于面积特征的直接分类器和基于傅立叶描述子的最小距离分类器对车辆进行自动分类。实验表明,本文的方法简单有效,可以获得较高的分类正确率。
This paper introduces an Automobile Automatic Recog-nition System based on image.It begins with the structures of the system.Then detailed methods for implementation are dis-cussed.Several methods,such as moving edge detection and converting to mono image,are integrated into the implentation for image pre-processing.Then charactersitics of the automobile are extracted by Fourier series.Finally automobiles are classified by area at first and then minimum distance classifier based on Fourier descriptors.The methods prove to be effective in the ex-periments and high validity can be obtained.
出处
《微计算机信息》
2003年第8期38-39,64,共3页
Control & Automation
