[R-sig-Geo] Announcement: new R package for satellite image time series analysis

[Gilberto Camara]

Dear R-SIG-GEO members

For those of you that are interested on satellite image time series
analysis, we would like to announce the availability of the beta version of
the R SITS package, which can be obtained from github:
> devtools::install_github("e-sensing/sits").

The SITS package is a set of tools for working with satellite image time
series. Includes data retrieval from a WTSS (web time series service),
different visualisation methods for image time series, smoothing methods
for noisy time series, different clustering methods  including dendrograms
and SOM. Matches noiseless patterns with noisy time series using the TWDTW
method for shape recognition. Provides machine learning methods for time
series classification, including SVM, LDA, QDA, GLM, Lasso, Random Forests
and Deep Learning.

The main purpose of SITS is to perform land use and land cover
classification using time series derived from satellites such as MODIS,
LANDSAT, and SENTINEL-2. It has been finetuned to deliver reasonable
performance for raster data sets in the GeoTIFF format.

The package has been developed by the team of the "e-sensing" project at
INPE (Brazil's National Institute for Space Research), supported by FAPESP
(Sao Paulo Research Foundation). We also received support from
International Climate Initiative of the Germany Federal Ministry for the
Environment, Nature Conservation, Building and Nuclear Safety.

We welcome comments, suggestions and bug reports on the package.  We
suggest that those who are interested take a look at the vignette, that can
be built using the command:
> devtools::build_vignettes()

Best regards
Gilberto
-- 
Prof. Dr. Dr. h.c. Gilberto Camara
National Institute for Space Research, INPE, Brazil
http://www.dpi.inpe.br/gilberto

	[[alternative HTML version deleted]]