节点文献
基于KL变换的汽车车型的识别方法
Vehicle Recognition Based on KL Transform
【摘要】 提出了一种基于KL变换的汽车车型的识别方法.首先对车辆图像进行预处理,然后通过KL变换构造特征车,将样本投影到特征车子空间,得到的投影系数即为待识别车型的代数特征,最后利用K近邻法进行车型识别.实验结果表明了文中提出的车辆识别方法是简便、快捷、有效的.
【Abstract】 This paper presents a kind of vehicle recognition method based on KL transform. Firstly, we preprocess the vehicle images. Then we build the eigen-vehicles by KL transform. The sample is projected onto the eigen-vehicle-subspace, and the coefficient of the projection is the algebra-feature of the vehicle. At last, we recognize the vehicle with KNN. Experimental results have shown good detective performance of the algorithm.
【关键词】 计算机科学与技术;
图像处理;
车型识别;
KL变换;
K近邻法;
【Key words】 computer science and technology; image processing; vehicle recognition; KL transform; KNN;
【Key words】 computer science and technology; image processing; vehicle recognition; KL transform; KNN;
- 【文献出处】 中南林学院学报 ,Journal of Central South Forestry University , 编辑部邮箱 ,2004年02期
- 【分类号】TP391.41
- 【被引频次】10
- 【下载频次】345