欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对智能车在视觉导航过程中车道线检测的鲁棒性和实时性问题,提出一种适用于结构化道路的车道线鲁棒检测与跟踪方法。首先,简化的Sobel算子提取车道线边缘图像,将边缘图像与改进的Otsu方法得到的车道线分割图像进行融合,实现对车道线标记点的鲁棒检测;然后,采用迭代最小二乘方法拟合车道线标记点并去除干扰点,并根据拟合参数建立车道线模型;最后,引入尺度无迹卡尔曼滤波（SUKF）对车道线进行跟踪。通过对多段实地采集的视频进行了仿真实验,结果表明,该方法对于高速公路车道线的检测率可达到99%,并具有较好实时性能;对于受损和弄污的城市道路车道线也体现出较好的鲁棒性和时间性能。
英文摘要：
Aiming at the robust and real time problems of lane detection in the visual navigation system of intelligent vehicles,a robust lane detection and tracking method is proposed for the structured road.Firstly,the lane marking pixels are reliably detected by fusing the lane edge image with the simplified Sobel algorithm and the lane segmentation image with the improved Otsu method.Secondly,the iterative least square method is proposed for fitting the lane markings and removing the non-lane markings.Then the lane model is constructed according to the fitting parameters.Finally,the scaled unscented Kalman filter（SUKF） is introduced for tracking the lanes and locating them in successive frames.The simulations of the proposed method are carried on several videos.The results show that the proposed method can supply the highway lane detection rate up to 99% and has good time performance.The algorithm is also applicable to impaired and dirty town roads,which also has better robust and time performance.
同期刊论文项目
同项目期刊论文
A spatial orthogonal allocation and heterogeneous cultural hybrid algorithm for multi-robot explorat
Vision-based long-distance lane perception and front vehicle location for full autonomous vehicles o
A regularity model-based multiobjective estimation of distribution algorithm with reducing redundant
Fault detection and identification for dead reckoning system of mobile robot based on fuzzy logic pa
Hybridizing particle swarm optimization with differential evolution for constrained numerical and en
期刊信息
位置：