elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Deep-Learning segmentation and 3D reconstruction of road markings using multi-view aerial imagery

Kurz, Franz and Azimi, Seyedmajid and Sheu, Chun-Yu and Angelo, Pablo (2019) Deep-Learning segmentation and 3D reconstruction of road markings using multi-view aerial imagery. ISPRS International Journal of Geo-Information, 8 (47), pp. 1-16. Multidisciplinary Digital Publishing Institute (MDPI). DOI: 10.3390/ijgi8010047 ISSN 2220-9964

[img] PDF
53MB

Official URL: http://www.mdpi.com/2220-9964/8/1/47/pdf

Abstract

The 3D information of road infrastructures are gaining importance with the development of autonomous driving. In this context, the exact 2D position of the road markings as well as the 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 apply a wavelet-enhanced fully convolutional network on multi-view 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 multi-view 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 show an improved IoU of the automatic road marking segmentation by exploiting the multi-view character of the aerial images and a more accurate 3D reconstruction of the road surface compared to the Semi Global 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.

Item URL in elib:https://elib.dlr.de/125720/
Document Type:Article
Title:Deep-Learning segmentation and 3D reconstruction of road markings using multi-view aerial imagery
Authors:
AuthorsInstitution or Email of AuthorsAuthors ORCID iD
Kurz, Franzfranz.kurz (at) dlr.dehttps://orcid.org/0000-0003-1718-0004
Azimi, SeyedmajidSeyedmajid.Azimi (at) dlr.dehttps://orcid.org/0000-0002-6084-2272
Sheu, Chun-Yuchun-yu.sheu (at) dlr.deUNSPECIFIED
Angelo, PabloPablo.Angelo (at) dlr.dehttps://orcid.org/0000-0001-8541-3856
Date:18 January 2019
Journal or Publication Title:ISPRS International Journal of Geo-Information
Refereed publication:Yes
Open Access:Yes
Gold Open Access:Yes
In SCOPUS:Yes
In ISI Web of Science:No
Volume:8
DOI :10.3390/ijgi8010047
Page Range:pp. 1-16
Publisher:Multidisciplinary Digital Publishing Institute (MDPI)
ISSN:2220-9964
Status:Published
Keywords:Aerial Image Sequences, Road marking detection, 3D Line-features Reconstruction, Fully Convolutional Neural Network
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Traffic Management (old)
DLR - Research area:Transport
DLR - Program:V VM - Verkehrsmanagement
DLR - Research theme (Project):V - Vabene++ (old)
Location: Oberpfaffenhofen
Institutes and Institutions:Remote Sensing Technology Institute > Photogrammetry and Image Analysis
Deposited By: Kurz, Dr.-Ing. Franz
Deposited On:10 Jan 2019 11:33
Last Modified:21 Sep 2019 05:05

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.