摘要
车道线的检测是汽车高级驾驶辅助系统中的重要组成部分。为了提高车道线检测的准确率,论文提出使用颜色阈值分割法检测车道线。首先对采集到的视频图像使用高斯滤波去噪,然后进行颜色空间转换,从RGB转换到HSV,再选取合适的颜色阈值对图像进行分割来实现目标识别(车道线)。目标区域运用Canny算子进行边缘检测,最后使用霍夫变换对车道线实时追踪。实验结果表明,该方法是可以实时对车道线检测,且准确率较高。
The detection of lane lines is an important part of the car's advanced driver assistance system.In order to improve the accuracy of lane line detection,this paper proposes the use of color threshold segmentation method to detect lane lines.First,Gaussian filtering is used to denoise the captured video image,then color space conversion is performed,and RGB is converted to HSV,and then the appropriate color threshold is selected to segment the image for target recognition(lane line).The Canny operator is used for edge detection in the target area,and finally the lane line detection is performed using the Hough transform.The experimental results show that the method can detect the lane line in real time with high accuracy.
作者
宋琴琴
杨国平
SONG Qinqin;YANG Guoping(School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science,Shanghai 201620)
出处
《计算机与数字工程》
2021年第9期1895-1898,共4页
Computer & Digital Engineering
