Division of Electronics and Electrical Engineering, Dongguk University, 30 Pildong-ro 1-gil, Jung-gu, Seoul 100-715, Korea.
Sensors (Basel). 2019 Jan 11;19(2):281. doi: 10.3390/s19020281.
Detection and classification of road markings are a prerequisite for operating autonomous vehicles. Although most studies have focused on the detection of road lane markings, the detection and classification of other road markings, such as arrows and bike markings, have not received much attention. Therefore, we propose a detection and classification method for various types of arrow markings and bike markings on the road in various complex environments using a one-stage deep convolutional neural network (CNN), called RetinaNet. We tested the proposed method in complex road scenarios with three open datasets captured by visible light camera sensors, namely the Malaga urban dataset, the Cambridge dataset, and the Daimler dataset on both a desktop computer and an NVIDIA Jetson TX2 embedded system. Experimental results obtained using the three open databases showed that the proposed RetinaNet-based method outperformed other methods for detection and classification of road markings in terms of both accuracy and processing time.
道路标记的检测和分类是自动驾驶车辆运行的前提。尽管大多数研究都集中在道路车道标记的检测上，但其他道路标记（如箭头和自行车标记）的检测和分类并没有受到太多关注。因此，我们提出了一种使用单阶段深度卷积神经网络（CNN），即 RetinaNet，在各种复杂环境下对道路上各种类型的箭头标记和自行车标记进行检测和分类的方法。我们在使用可见光相机传感器捕获的三个公开数据集的复杂道路场景中对所提出的方法进行了测试，这三个数据集分别是 Malaga 城市数据集、Cambridge 数据集和 Daimler 数据集，测试平台包括桌面计算机和 NVIDIA Jetson TX2 嵌入式系统。使用这三个公开数据库获得的实验结果表明，所提出的基于 RetinaNet 的方法在道路标记的检测和分类方面在准确性和处理时间方面都优于其他方法。
