摘要
针对车辆的侧面图像,设计了一种基于数字图像处理技术的车辆图像特征参数识别方法。为克服噪声的影响,采用自适应高斯混合模型进行背景抽取,在此基础上对车长和车高的测量方法进行了研究,通过实验分析了车速和物距对测量精度的影响。试验结果表明:在低速范围内车速对测量精度的影响很小,物距对测量结果的影响较大。在此基础上建立了车辆外形参数与物距的关系,可有效测量车辆的外形参数,为进一步对车辆进行识别分类提供了参考依据。
For vehicle's side image,the method of identifying body characteristics is designed on digital image processing technology.To reduce the effect from noise,a Gaussian mixture model(GMM)is adopted to abstract the background.The measure method on body length and body height is studied.By experiments,the effects of vehicle speed and object distances on measure accuracy are analyzed.The result shows that vehicle speed affects the measure accuracy little while object distance affects comparatively much in the range of low speed.Based on that,the correlation between body parameter and object distance is found,and body parameter is measured validly.Those are the consultative basis for further study on identification and classification of vehicle.
出处
《河南科技大学学报（自然科学版）》
CAS
2007年第3期17-20,共4页
Journal of Henan University of Science And Technology:Natural Science
基金
山东省自然科学基金项目(Q99F11)
关键词
车辆外形参数
自适应背景抽取
车速
物距
Vehicle body parameter
Adaptive background abstracting
Vehicle speed
Object distance
