复杂交通场景中采用稀疏表示的车辆识别方法.pdfVIP

第 17 卷第3 期 中国图象图形学报 Vol. 17 , No.3 2012 年3 月 Joumal of Image and Graphics Mar. , 2012 中图法分类号:1凹91. 4 文献标志码:A 文章编号: 1∞6-8961(2012)03-0387-06 论文引用格式:李修志，吴健，崔志明，陈建明.复杂交通场景中采用稀疏表示的车辆识别方法[J] .中国图象图形学报，2012 ， 17(3) :387-392 复杂交通场景中采用稀疏表示的车辆识别方法 李修志1 ，吴健1 气崔志明1 气陈建明1 ， 2 1.苏州大学智能信息处理及应用研究所，苏州 215∞6; 2. 江苏怡和科技股份有限公司，苏州 215∞2 摘 要:针对智能交通系统中复杂交通场景下对于车辆识别的难点问题，提出一种基于方向梯度直方图特征和稀 疏表示的车辆目标识别方法。该方法首先使用方向梯度直方图特征分别提取训练样本和待测目标信息，通过稀疏 表示的方法将训练样本训练为过完备字典，最后通过待测目标在字典中稀疏度和重构残差判定目标是否为车辆并 对其标注。实验结果表明，提出的方法在粘连遮挡、目标类别多样等复杂交通场景中有较好的识别率和实时性。 关键词:稀疏表示;方向梯度直方图;车辆识别;智能交通;压缩感知 Sparse representation method of vehicle recognition in complex traffic scenes Li Xiuzhi1 ,Wu Jiar111 ,Cui aiming1 ,2 ,Cherl Jianmint ,2 1. The Instit脚01 InteUigent b协rmation Processing and Application ,S附加即 Univers仰， Suzhou 215∞6 ， China; 2. Jiangsu yi品e Technology Co. ,Ltd. ,Suzhou 215∞2 ， China AbstJ:田t: For intelligent transportation systems , vehicle recognition in complex traffic scenes is a key issue. In this article , a novel scheme using (HOG) features and sparse representation target recognition for vehicle recognition in complex traffic scenes is proposed. Our method uses the HOG to extract features from samples and candidate targets , and then wses trained samples as an overcomplete dictionary based on sp町se representation. Finally , candidate targets are recognized by computing sparsity and reconstruction residuals in tbe dictionary. Experiment results show that tbe proposed scheme provides higher recognition preciseness in real time , even in complex traffic scenes such containi