Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회지)

Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회)
권호 표지
DOI QR코드

DOI QR Code

Automatic Drawing and Structural Editing of Road Lane Markings for High-Definition Road Maps

정밀도로지도 제작을 위한 도로 노면선 표시의 자동 도화 및 구조화

  • Choi, In Ha (Department of Spatial Information Engineering, Namseoul University) ;
  • Kim, Eui Myoung (Department of Drone.GIS Engineering, Namseoul University)
  • Received : 2021.10.18
  • Accepted : 2021.11.29
  • Published : 2021.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

High-definition road maps are used as the basic infrastructure for autonomous vehicles, so the latest road information must be quickly reflected. However, the current drawing and structural editing process of high-definition road maps are manually performed. In addition, it takes the longest time to generate road lanes, which are the main construction targets. In this study, the point cloud of the road lane markings, in which color types(white, blue, and yellow) were predicted through the PointNet model pre-trained in previous studies, were used as input data. Based on the point cloud, this study proposed a methodology for automatically drawing and structural editing of the layer of road lane markings. To verify the usability of the 3D vector data constructed through the proposed methodology, the accuracy was analyzed according to the quality inspection criteria of high-definition road maps. In the positional accuracy test of the vector data, the RMSE (Root Mean Square Error) for horizontal and vertical errors were within 0.1m to verify suitability. In the structural editing accuracy test of the vector data, the structural editing accuracy of the road lane markings type and kind were 88.235%, respectively, and the usability was verified. Therefore, it was found that the methodology proposed in this study can efficiently construct vector data of road lanes for high-definition road maps.

정밀도로지도는 자율주행차의 기본 인프라로 활용되어 최신 도로정보가 신속하게 반영되어야 한다. 하지만 현재 정밀도로지도 공정 중 객체 도화 및 구조화 편집과정이 수작업으로 이루어지며 주요 구축 대상인 도로 노면선 표시의 레이어를 생성하는데 가장 오랜 시간이 소요된다. 이에 본 연구에서는 선행 연구에서 기학습된 포인트넷(PointNet) 모델을 통해 색상 유형(백색, 청색, 황색)이 예측된 도로 노면선 표시의 포인트 클라우드를 입력 데이터로 활용하였고, 이를 기반으로 본 연구에서는 도로 노면선 표시 레이어의 도화 및 구조화 편집을 자동화하는 방법론을 제안하였다. 제안한 방법론을 통해 구축한 3차원 벡터 데이터의 활용성을 검증하기 위해 정밀도로지도 품질검사 기준에 따라 정확도를 분석하였다. 벡터 데이터의 위치정확도 검사에서 수평 오차와 수직 오차에 대한 평균제곱근오차(RMSE: Root Mean Square Error)는 0.1m 이내로 나타나 적합성을 검증하였으며, 구조화 편집 정확도 검사에서 선표시 유형과 선규제 유형의 구조화 정확도가 모두 88.235%로 나타나 활용성을 검증하였다. 따라서, 본 연구에서 제안한 방법론으로 정밀도로지도를 위한 도로 노면선 표시의 벡터 데이터를 효율적으로 구축할 수 있는 것을 알 수 있었다.

Keywords

Acknowledgement

이 논문은 2020년도 정부(국토교통부)의 재원으로 공간정보 융복합 핵심인재 양성사업의 지원을 받아 수행된 연구임(2020-02-01).

References

  1. National Geographic Information Institute (2020), Manual of quality inspection criteria of the high-definition road maps, pp. 1-136.
  2. National Geographic Information Institute (2021), Explanation and guidance materials of high-definition road maps, pp. 1-51.
  3. Choi, H.K. (2021), A study and suggestion of digital twin realization in korean new deal, Review of Architecture and Building Science, Vol. 65, No. 3, pp. 38-42.
  4. Choi, I.H. and Kim, E.M. (2021), Automatic construction of deep learning training data for high-definition road maps using mobile mapping system, The Journal of Korea Society of Surveying, Geodesy, Photogrammetry, and Cartography, Vol. 39, No. 3, pp. 133-139. (in Korean with English abstract) https://doi.org/10.7848/KSGPC.2021.39.3.133
  5. Choi, K.J. and Kim, K.H. (2021), An automatic extraction method for road maps using LiDAR-based 3D point maps, The Korean Institute of Information Scientists and Engineers, Vol. 27, No. 5, pp. 234-240. (in Korean with English abstract)
  6. Choi, T.S., Yoon, H.S., Choi, Y.S., Lee, W.J., and Chang, S.Y (2020), A study on high definition road map construction using aerial photography, The Journal of Korean Society for Geospatial Information Science, Vol. 28, No. 3, pp. 69-76. (in Korean with English abstract)
  7. Korean Society of Transportation (2021), Trend of Policy, Transportation Technology and Policy, Vol. 18, No. 2, pp. 90-92.
  8. Qi, C.R, Su, H., Mo, K., and Gulibas, L.J (2017), PointNet: deep learning on point sets for 3D classification and segmentation, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 652-660.