Large-Scale 3D Roadside Modelling with Road Geometry Analysis: Digital Roads New Zealand

Latest developments in camera technology and computer vision, as well as in computer and communication technologies, contribute to improving safety on roads by the development of new key technologies, such as autonomous driving or driver-assistance systems. These new technologies need to be tested extensively and purposefully, and supported by environment or infrastructure models or integrated sensors. The paper reports about a project "Digital Roads New Zealand". Novel sensor technologies, including stereo vision and odometry, have been used for recording, modelling and analysing a large test site. It is demonstrated how those data can be used for detecting changes in road geometry, such as various forms of road surface distress. An important novelty of the shown solution is the scale of the project (i.e. size of the digitised area using car-mounted sensors) together with the achieved very high accuracy in road-geometry analysis.

[1]  Dariu Gavrila,et al.  The Issues , 2011 .

[2]  Gerd Hirzinger,et al.  View Planning for Multi-View Stereo 3D Reconstruction Using an Autonomous Multicopter , 2012, J. Intell. Robotic Syst..

[3]  Reinhard Klette,et al.  Road surface distress detection in disparity space , 2017, 2017 International Conference on Image and Vision Computing New Zealand (IVCNZ).

[4]  Darius Burschka,et al.  Adaptive and Generic Corner Detection Based on the Accelerated Segment Test , 2010, ECCV.

[5]  Reinhard Klette.  Vision-Based Driver Assistance , 2015 .

[6]  Reinhard Klette,et al.  Concise Computer Vision , 2014, Undergraduate Topics in Computer Science.

[7]  Jürgen Wohlfeil,et al.  EXTENSION AND EVALUATION OF THE AGAST FEATURE DETECTOR , 2016 .

[8]  Heiko Hirschmüller,et al.  Evaluation of Stereo Matching Costs on Images with Radiometric Differences , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  P. Burrough,et al.  Principles of geographical information systems , 1998 .

[10]  Jürgen Wohlfeil,et al.  Eligible Features Segregation for Real-time Visual Odometry , 2017 .

[11]  Peter S. Maybeck,et al.  Stochastic Models, Estimation And Control , 2012 .

[12]  Reinhard Klette,et al.  Landmark Initialization for Unscented Kalman Filter Sensor Fusion in Monocular Camera Localization , 2013, Int. J. Fuzzy Log. Intell. Syst..

[13]  Denis Grießbach.  Stereo-Vision-Aided Inertial Navigation , 2015 .

[14]  Dirk Baumbach,et al.  Stereo-vision-aided inertial navigation for unknown indoor and outdoor environments , 2014, 2014 International Conference on Indoor Positioning and Indoor Navigation (IPIN).

[15]  P. Burrough.  Principles of Geographical Information Systems for Land Resources Assessment , 1986 .

[16]  Waqar A. Khan,et al.  Accuracy of Stereo-Based Object Tracking in a Driver Assistance Context , 2013 .

[17]  Adrian Schischmanow,et al.  IPS – a vision aided navigation system , 2017 .

[18]  Jürgen Wohlfeil,et al.  Uncertainty Model for Template Feature Matching , 2017, PSIVT.

[19]  Reinhard Klette,et al.  Panoramic Imaging: Sensor-Line Cameras and Laser Range-Finders , 2008 .

[20]  Ines Ernst,et al.  Mutual Information Based Semi-Global Stereo Matching on the GPU , 2008, ISVC.