提示
基于路径搜索的道路检测算法
首发时间：2023-02-10
摘要：针对有车道线和无车道线的结构化道路检测方法，提出一种基于路径搜索的道路检测算法。首先选择感兴趣区域减少计算量，采用途像素分割算法提取车道线起始搜索点并确定搜索框，基于梯度、像素、纹理变化进行特征融合，根据特征变化大小从起始搜索框出发进行搜索，最后对选取点进行曲线拟合形成车道线。该算法相比于经典hough变换道路检测具有鲁棒性更强、在弯曲道路线检测效果更好的特点，在有无车道线的结构化路面都表现良好。
关键词：
for information in english, please click here
road detection algorithm based on path search
abstract：aiming at the structured road detection method with lane lines and no lane lines, a road detection algorithm based on route search is proposed. firstly, the area of interest is selected to reduce the amount of calculation.the pixel segmentation algorithm is used to extract the starting search point of the lane line and determine the search box. the feature fusion is carried out according to the gradient, pixel and texture changes.the search is carried out from the starting search box according to the size of the feature change.finally the selected point is curved fitting to form the lane line. compared with the road detection of the hough transform, the algorithm has the characteristics of more robustness and better performance in curved road line.and the algorithm shows good results in structured road surfaces with or without lane lines.
keywords：
基金：
论文图表：
引用
导出参考文献
no.****
动态公开评议
共计0人参与
勘误表
基于路径搜索的道路检测算法
评论
全部评论