【作者】 邓天民； 邵毅明； 崔建江；

【Author】 DENG Tianmin SHAO Yiming CUI Jianjiang school of traffic & transportation, ChongQing Jiaotong University, ChongQing, 400074 school of science & information engineering, North East University, ShenYang, 110004

【机构】 重庆交通大学交通运输学院； 东北大学科学与信息工程学院；

【摘要】 提出了一种基于Gabor小波变换和隐马尔可夫模型的车型识别算法。该算法首先对图像进行预处理,然后通过Gabor小波变换与主元分析法构造Gabor特征车辆,最后将Gabor特征车辆的特征节做为观测向量,进行隐马尔可夫模型训练与优化模型参数。实验结果表明了文中提出的车型识别方法是简便、快捷、有效的。更多还原

【Abstract】 A vehicle recognition algorithm based on Gabor wavelets transform and hidden Markov model(HMM) is proposed. Firstly, the vehicle images is pre-processed. Then, the Gabor-Vehicle is build by Gabor filter and principal component analysis (PCA). At last, a set of images representing different instances of the same vehicle are used to train each HMM, and each individual in the database is represented by an optional HMM vehicle model. Experimental results have shown good detective performance of the algorithm.更多还原