基于CCD摄像技术的公交车跟车距离监测系统研究

为了实时有效地控制公交车行驶时的安全测距,以Matlab工具箱函数为基础,运用图像灰度化处理、图像去噪、图像二值化,从处理后的图像中提取出具有表征意义的统计特征。依据预先设定的图像模板,通过比对所提取的图像特征和模板图像的特征,利用图像匹配技术进行图像识别。在此基础上建立公交车跟车距离识别模型。采用实车实验数据对模型进行验证,结果表明:在静态情况下识别距离与真实距离之间的误差总体上随着距离的增加而逐渐加大,平均误差为5%,绝大部分偏差在6%以下。动态情况下验证误差较静态大,均值为7.68%,标准偏差为0.012 56,且随着车速的增加,真实值与识别值之间的绝对误差相应变大。

Research on car-following distance detection system of bus based on CCD camera technology

In order to control the safe distance of the bus in real time, based on the Matlab toolbox function, we take the Image gray processing, image denoising processing, image binarization processing. The statistical characteristics of the image are extracted from the processed images. By comparing the image features extracted from the image and template image, the image matching technique is used to carry out the image recognition. The model is verified by the experimental data of real vehicle.The results show that: in the static condition, the error between the recognition distance and the real distance increases gradually, and the average error is 5%, the majority of the error is below 6%. In dynamic case, the error is bigger, the mean is 7.68%, the standard deviation is 0.01256, and the absolute error between the real value and the value of the identification value increases with the vehicle speed.