%0 Conference Proceedings
%4 sid.inpe.br/marte2/2017/10.27.12.51.50
%2 sid.inpe.br/marte2/2017/10.27.12.51.51
%@isbn 978-85-17-00088-1
%F 59526
%T Classificação de feições na superfície a partir de dados LiDAR e medidas de entropia e desvio padrão das altitudes
%D 2017
%A Oliveira, Renan Américo Ribeiro de,
%A Galo, Mauricio,
%@electronicmailaddress renanamerico@gmail.com
%E Gherardi, Douglas Francisco Marcolino,
%E Aragão, Luiz Eduardo Oliveira e Cruz de,
%B Simpósio Brasileiro de Sensoriamento Remoto, 18 (SBSR)
%C Santos
%8 28-31 maio 2017
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 2892-2899
%S Anais
%1 Instituto Nacional de Pesquisas Espaciais (INPE)
%X This paper describes a methodological approach to automatic classification of LiDAR data aiming at filtering and classifying the following classes: ground, buildings, vegetation and transmission lines. The classification approach is based on the use of the following attributes: height standard deviation and height entropy, among others attributes, estimated from LiDAR point cloud. Those attributes enable to find patterns in this kind of data through the establishment of limits on normal distribution curves, which later serve as parameters for sequentially filtering and classifying in the classes mentioned. Besides these attributes, some additional filters, such as the minimum height to distinguish building points, removal of isolated points and coherence analysis of height between nearby objects, were used to refine the classification at the end of the algorithm. The results obtained by the proposed algorithm were compared with the results obtained by other software. From a visual analysis of these two classifications and from the optical image of the same area, it was possible to highlight the positive and negative aspects of the proposed approach. From these results it was verified that, despite the need for some improvements, it was possible to perform the classification based on attributes such as entropy and standard deviation estimated from the point cloud obtained by LiDAR systems.
%9 LIDAR: sensores e aplicações
%@language pt
%3 59526.pdf