%0 Conference Proceedings
%4 sid.inpe.br/mtc-m16c/2018/12.27.18.12
%2 sid.inpe.br/mtc-m16c/2018/12.27.18.12.05
%@issn 2179-4847
%T SP-TWDTW: a new parallel algorithm for spatio-temporal analysis of remote sensing images
%D 2018
%A Oliveira, Savio S. T. de,
%A Pascoal, Luiz M. L.,
%A Ferreira, Laerte,
%A Cardoso, Marcelo de C.,
%A Bueno, Elivelton,
%A Rodrigues, Vagner J. S.,
%A Martins, Wellington S.,
%@affiliation Universidade Federal de Goiás (UFG)
%@affiliation Universidade Federal de Goiás (UFG)
%@affiliation Universidade Federal de Goiás (UFG)
%@affiliation Universidade Federal de Goiás (UFG)
%@affiliation Universidade Federal de Goiás (UFG)
%@affiliation Universidade Federal de Goiás (UFG)
%@affiliation Universidade Federal de Goiás (UFG)
%E Vinhas, Lúbia (INPE),
%E Campelo, Claudio (UFCG),
%B Simpósio Brasileiro de Geoinformática, 19 (GEOINFO)
%C Campina Grande
%8 05-07 dez. 2018
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 46-57
%X In the class of computationally complex problems, the time series analysis is one of those that has high demand for computational power. The Time-Weighted Dynamic Time Warping (TWDTW) algorithm stands out as one of the best solution found in the literature in this field, but its time complexity of O(n2) makes it unfeasible for large data sets. To overcome this limitation, this work proposes a parallel algorithm, named SP-TWDTW (Spatial Parallel TWDTW), that allows the analysis of large scale time series using Manycore architectures. The SP-TWDTW considers the temporal axis and the spatial au- tocorrelation to determine the land use mapping in a given region. The results show that the SP-TWDTW algorithm is a promising solution with response time up to 11 times lower.
%@language pt
%3 p5.pdf