Road markings are the primary road feature of High-definition(HD) map in automatic vehicle and play a critical role in traffic safety. Mostly methods are sensitive to intensity value of laser data and the type of road scene. To solve this problem, this paper proposed a method using the image-to-image translation based on conditional adversarial networks to achieve the extraction, recognition and identification of road marking based on laser data. This method contains three steps: (1) automatic filter of ground surface based on topological network framework of laser point, (2) an automated road marking extraction and recognition conditional adversarial networks of image-to-image method, (3) the identification and vectorization based on KD tree clustering algorithm. Quantitative and qualitative analysis based on experimental data for different road scenarios were used to verify the robustness of the method and the accuracy of the extraction results. The experimental result based on different road scenes is promising and valuable for the update of road feature database. The proposed method makes the extraction, recognition, and identification of road markings more efficient and accurate and delivers a valuable solution for the HD map of intelligent and connected vehicle.

CICTP