Abstract

Abstract. The city management is increasingly supported by information technologies, leading to paradigms such as smart cities, where decision-makers, companies and citizens are continuously interconnected. 3D modelling turns of great relevance when the city has to be managed making use of geospatial databases or Geographic Information Systems. On the other hand, laser scanning technology has experienced a significant growth in the last years, and particularly, terrestrial mobile laser scanning platforms are being more and more used with inventory purposes in both cities and road environments. Consequently, large datasets are available to produce the geometric basis for the city model; however, this data is not directly exploitable by management systems constraining the implementation of the technology for such applications. This paper presents a new algorithm for the automatic detection of zebra crossing. The algorithm is divided in three main steps: road segmentation (based on a PCA analysis of the points contained in each cycle of collected by a mobile laser system), rasterization (conversion of the point cloud to a raster image coloured as a function of intensity data), and zebra crossing detection (using the Hough Transform and logical constrains for line classification). After evaluating different datasets collected in three cities located in Northwest Spain (comprising 25 strips with 30 visible zebra crossings) a completeness of 83% was achieved.

Highlights

  • Cities are increasingly large and complex areas that require integrated technologies for an effective management and to ensure productivity, continued economic growth and environmental sustainability

  • The city management is increasingly supported by information technologies, leading to paradigms such as smart cities, where decision-makers, companies and citizens are continuously interconnected. 3D modelling turns of great relevance when the city has to be managed making use of geospatial databases or Geographic Information Systems

  • Some zebra crossing were not detected, false positives are not presented in the results

Read more

Summary

Introduction

Cities are increasingly large and complex areas that require integrated technologies for an effective management and to ensure productivity, continued economic growth and environmental sustainability. 3D modelling turns of great relevance when the city has to be managed making use of geospatial databases or Geographic Information Systems. In this sense, the new concepts of 3D modelling not accounting only for geometry and for semantic and topologic data became of essential importance. The new concepts of 3D modelling not accounting only for geometry and for semantic and topologic data became of essential importance Standard schemas such as cityGML contribute to define how 3D models need to be structured in order to be used under an interoperable perspective. Large datasets are available to produce the geometric basis for the city model, this data is not directly exploitable by management systems constraining the implementation of the technology for such applications

Methods
Results
Conclusion
Full Text
Published version (Free)

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 call