High-Accuracy Lane Geodetic Coordinates Extraction Based on Camera-Lidar Fusion

For various transportation demands, map application has become indispensable in people's lives. In map applications, map data affects the navigation engine to resolve the optimum travel time and provides the road geometry information of the surrounding environment for users. Therefore, high accuracy with frequent update map data is desired. This work presents a workflow to extract the accurate lane geodetic coordinates automatically. According to the experiment results, the proposed approach shows that lane lines can be extracted and secured with sub-meter accuracy by integrating commercial-grade cameras and non-survey-grade Lidar. Thus, the cost-effective structure for automatic map creation is firmed. It will further benefit geodetic object extraction usage.