基于深度学习的无人驾驶汽车车道线检测方法

车道线检测是实现当前汽车辅助驾驶和未来无人驾驶汽车的关键,深度学习技术在近年来迅猛发展,在图像识别、图像分割、语音识别及数据预测等方面都取得了出色成绩。结合深度学习技术对无人驾驶汽车环境感知中的车道线检测进行了相应的研究,提出一种基于深度学习的车道线识别算法。对比研究已有算法,针对其中的信息融合问题,提出了一种新的特征图上下文信息融合方法,将该方法与VGG(Visual Geometry Group)网络相结合提出融合上下文信息的车道线识别网络VGG-FF,进一步加入空洞卷积提出融合空洞卷积及上下文信息的车道线识别网络VGG-FFD。将该网络模型在公开数据集以及自制数据集上进行了性能测试,实验结果表明该模型具有良好的识别效果。

Research on Lane Line Detection Method of

Lane line detection is the key to the realization of current car-assisted driving and future driverless cars. Deep learning technology has developed rapidly in recent years and has achieved excellent results in image recognition, image segmentation, speech recognition, and data prediction. This paper combines deep learning technology to carry out corresponding research on lane line detection in driverless car environment perception and proposes a lane line recognition algorithm based on deep learning. Comparative study of existing algorithms, given the information fusion problem, the paper proposes a novel feature map context information fusion method. The method is combined with the VGG network to propose VGG-FF, and the dilated convolution is further added to propose VGG-FFD lanes Line recognition network. The network model has been compared with the public data set and self-made data set. The experimental results prove that the model has a good recognition effect.