Lane Detection Base on Deep Learning
2018 11th International Symposium on Computational Intelligence and Design (ISCID)(2018)
摘要
In recent years, the automatic driving technology has developed rapidly. The detection of the lane line is one of the important contents. This paper sorts out the recent lane detection algorithm and the deep learning network, and analyzes the network VPGNet based on the segmentation to detect the lane line, and using the object detection-based algorithm RFCN for experimental comparison. According to the analysis of the experimental results, the algorithm based on object detection has advantages in object localization, but it still needs to use post-processing to fit a straight line. In terms of accuracy, VPGNet is higher than RFCN.
更多查看译文
关键词
object detection,deep learning,lane detection,VPGNet,RFCN
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络