摘要
本文提出一种基于图像处理的新型车辆距离检测技术,以解决传统超声波车距检测误差大、盲区多等问题。采用同一参数的三目摄像头标定技术,运用区域匹配算法寻找共轭测量点,既能有效减少后期数据计算工作量,提升算法运行速度,又能减少目标测量盲区。利用双目视觉原理,建立距离测量矩阵库,并实时更新矩阵库模型,提高距离测量精度。实验结果表明,这个方法能较高精度检测出不同角度、不同位置目标间距离。
In this paper,a new vehicle distance detection technology based on image processing is proposed to solve the problems of large error and many blind areas.Using the same parameters of three cameras calibration technology and the region matching algorithm,we can find the conjugate measurement points.This method not only reduces the post-data calculation workload and blind area effectively,but also improves the algorithm running speed.The distance measurement matrix library is established by using the binocular vision principle.The matrix library model is updated in real time to improve the distance measurement accuracy.Experimental results show that this method can detect the distance accurately under different angles and targets in different locations conditions.
作者
张雪
Zhang Xue(School of Information Engineering,Baise College,Baise Guangxi 533000,China)
出处
《信息与电脑》
2018年第21期108-110,共3页
Information & Computer
关键词
图像处理
车距检测
测量盲区
矩阵库
image processing
distance detection
measurement blind zone
matrix library
