%0 Conference Proceedings
%4 sid.inpe.br/mtc-m18/2013/12.11.12.08
%2 sid.inpe.br/mtc-m18/2013/12.11.12.08.57
%@issn 2179-4820
%T A method to automatically identify road centerlines from georeferenced smartphone data
%D 2013
%A Costa, George Henrique Rangel,
%A Baldo, Fabiano,
%@affiliation Universidade do Estado de Santa Catarina (UDESC)
%@affiliation Universidade do Estado de Santa Catarina (UDESC)
%E Andrade, Pedro Ribeiro,
%E Santanchè, André,
%B Simpósio Brasileiro de Geoinformática, 14 (GEOINFO).
%C Campos do Jordão
%8 24-27 nov. 2013
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 12
%S Anais
%X Digital road maps have gained fundamental role in populations daily life, so they need to be accurate and up-to-date. A viable solution is to generate maps by processing GPS data. However, one of the most challenging tasks regarding this approach is how to extract road centerlines from the cloud of georeferenced GPS points. The literature presents various methods that do it, but none have been found that is prepared for the continuous update of the map and refinement of its roads accuracy. In this context, the objective of this work is to propose a method to identify road centerlines using an evolutive algorithm in order to generate and update road maps. This work uses as source of data GPS traces collected by smartphones. Although suitable to obtain a large amount of georeferenced data, these devices bring the additional problem of identifying the transport vehicle used along each trace. Preliminary results indicate that the methods performance is satisfactory, with an average variation of 2.95 meters in relation to satellite images.
%@language en
%3 paper8.pdf