Automatic Lane Marking Extraction From Point Cloud Into Polygon Map Layer

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F17%3A43911676" target="_blank" >RIV/62156489:43110/17:43911676 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/00216305:26210/17:PU124412

Výsledek na webu

<a href="http://symposium.earsel.org/37th-symposium-Prague/wp-content/uploads/2016/06/2017_EARSeL_abstract_book.pdf" target="_blank" >http://symposium.earsel.org/37th-symposium-Prague/wp-content/uploads/2016/06/2017_EARSeL_abstract_book.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Automatic Lane Marking Extraction From Point Cloud Into Polygon Map Layer

Popis výsledku v původním jazyce

Optimization of road networks is a common concern worldwide, primarily for safety purposes. Because the extent of these networks is substantial, automation of their inventory is highly desirable. This paper concentrates on the road inventory process that is necessary for regular maintenance. The key part of our road marking detection and reconstruction is based on spanning tree usage. The spanning trees are obtained from alpha shapes of the detected road markings. Application of the spanning trees enables the reliable identification of the road markings and precise reconstruction of the of their contours even with noisy data. Our method processes the point cloud data obtained from LiDAR measurements, and provides a common vector layer with road line polygons. Such a vector layer is stored in a common file format ESRI Shapefile supported by the majority of geographical information systems, thus producing an output that can be conveniently used for decision-making based on the road inventory process. Our key advantage is the ability to reconstruct the precise shapes of the road lane markings. The reconstructed shapes can be used for the visualization as well as for analytic purposes. Our approach is very reliable as shown in the results. We used two different testing areas. The first tested area is located in the Brno city centre in the Czech Republic, particularly in the vicinity of the Mendel Square. It comprises common streets and squares. The roads contain multiple road lanes for cars as well as separate lanes for trams and trolleybuses. The second tested area is the country round near Semice town in the Czech Republic. The true positive rate of detection is in all cases above 92 %. Therefore, we are able to detect correctly vast majority of road lane markings.

Název v anglickém jazyce

Automatic Lane Marking Extraction From Point Cloud Into Polygon Map Layer

Popis výsledku anglicky

Optimization of road networks is a common concern worldwide, primarily for safety purposes. Because the extent of these networks is substantial, automation of their inventory is highly desirable. This paper concentrates on the road inventory process that is necessary for regular maintenance. The key part of our road marking detection and reconstruction is based on spanning tree usage. The spanning trees are obtained from alpha shapes of the detected road markings. Application of the spanning trees enables the reliable identification of the road markings and precise reconstruction of the of their contours even with noisy data. Our method processes the point cloud data obtained from LiDAR measurements, and provides a common vector layer with road line polygons. Such a vector layer is stored in a common file format ESRI Shapefile supported by the majority of geographical information systems, thus producing an output that can be conveniently used for decision-making based on the road inventory process. Our key advantage is the ability to reconstruct the precise shapes of the road lane markings. The reconstructed shapes can be used for the visualization as well as for analytic purposes. Our approach is very reliable as shown in the results. We used two different testing areas. The first tested area is located in the Brno city centre in the Czech Republic, particularly in the vicinity of the Mendel Square. It comprises common streets and squares. The roads contain multiple road lanes for cars as well as separate lanes for trams and trolleybuses. The second tested area is the country round near Semice town in the Czech Republic. The true positive rate of detection is in all cases above 92 %. Therefore, we are able to detect correctly vast majority of road lane markings.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů