中国计量大学学报2017，Vol.28Issue(4)：460-466,7.DOI:10.3969/j.issn.2096-2835.2017.04.009
结合Hough变换与Kalman滤波的车道线检测方法
Lane detection method combining Hough transform with Kalman filtering
郭笙听 1李子印 1赵储1
作者信息
- 1. 中国计量大学光学与电子科技学院,浙江杭州310018
- 折叠
摘要
Abstract
Lane detection,which uses the vehicle-mounted camera to get road image information,plays an important role in intelligent vehicle driving systems.An effective lane detection method was proposed to improve the accuracy and robustness of visual navigation.First,the original RGB image was cut to get the region of interest and processed with inverse perspective transformation,graying and threshold.Then,the lines in ROI(region of interest)were detected by Hough transform according to the position of center points and the line feature in image axis.The kalman filter algorithm was used to estimate the target position.The algorithm could effectively detect indistinct lanes and remove some interference in the image.The accuracy of the algorithm was up to 94%.And the algorithm had good reliability,robustness,which was conducive to the construction of real-time detection and tracking systems.关键词
车道线检测/逆透视变换/霍夫变换/卡尔曼滤波Key words
lane detection/inverse perspective transformation/Hough transform/Kalman filter分类
信息技术与安全科学引用本文复制引用
郭笙听,李子印,赵储..结合Hough变换与Kalman滤波的车道线检测方法[J].中国计量大学学报,2017,28(4):460-466,7.
