Abstract
The 3D information of road infrastructures is growing in importance with the development of autonomous driving. In this context, the exact 2D position of road markings as well as height information play an important role in, e.g., lane-accurate self-localization of autonomous vehicles. In this paper, the overall task is divided into an automatic segmentation followed by a refined 3D reconstruction. For the segmentation task, we applied a wavelet-enhanced fully convolutional network on multiview high-resolution aerial imagery. Based on the resulting 2D segments in the original images, we propose a successive workflow for the 3D reconstruction of road markings based on a least-squares line-fitting in multiview imagery. The 3D reconstruction exploits the line character of road markings with the aim to optimize the best 3D line location by minimizing the distance from its back projection to the detected 2D line in all the covering images. Results showed an improved IoU of the automatic road marking segmentation by exploiting the multiview character of the aerial images and a more accurate 3D reconstruction of the road surface compared to the semiglobal matching (SGM) algorithm. Further, the approach avoids the matching problem in non-textured image parts and is not limited to lines of finite length. In this paper, the approach is presented and validated on several aerial image data sets covering different scenarios like motorways and urban regions.
Highlights
The availability of large-scale, accurate high-resolution 3D information of roads with lane markings and road infrastructure plays an important role towards autonomous driving
Based on the resulting 2D segments in the original images, we propose a successive workflow for the 3D reconstruction of road markings based on a least-squares line-fitting in multiview imagery
The maximum number of contributing images depends on the flight configuration and is only reached in some parts, i.e., in the experiments, the corresponding road marking was segmented in eight contributing images
Summary
The availability of large-scale, accurate high-resolution 3D information of roads with lane markings and road infrastructure plays an important role towards autonomous driving. Aerial imagery is a valuable database to derive 3D information of roads even in areas difficult to access, like on motorways. Driven by the development of autonomous driving, area-wide, high-resolution 3D information of the road surfaces, including lane markings, is necessary, as well as new methods to derive this information from aerial imagery, as shown in Reference [1]. The lane markings, for example, are the most visible texture on asphalt roads useful for 3D reconstruction. It is desired to improve the quality of the DSM on road surfaces by exploiting the line character of the lane markings
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
