Provides an implementation of the Time-Weighted Dynamic Time Warping (TWDTW) method for land cover mapping using satellite image time series. TWDTW is based on the Dynamic Time Warping technique and has achieved high accuracy for land cover classification using satellite data. The method is based on comparing unclassified satellite image time series with a set of known temporal patterns (e.g. phenological cycles associated with the vegetation). Using 'dtwSat' the user can build temporal patterns for land cover types, apply the TWDTW analysis for satellite datasets, visualize the results of the time series analysis, produce land cover maps, create temporal plots for land cover change, and compute accuracy assessment metrics.
Adds dtwSat paper published on Journal of Statistical Software
Fix error in plotAccuracy
Generalizes twdwAssess to cases with only one map
Fixes error in getTimeSeries due to time series with only one no observation
Include the function twdtwApplyParallel for TWDTW parallel processing using the package snow
Include writeRaster for twdtwRaster class
Improve tests and documentation
Improve memory usage of twdtwApply
Improve memory usage and speed of twdtwClassify
Auto recognition of the argument "doy" to avoid naming the argument "doy = doy"
Fix bug in twdtwAssess for class twdtwMatches
Fix bug in twdtwRaster
Register TWDTW as a distance function into package proxy
Fix typos in plot labels
New accuracy metrics (twdtwAssess) for classified map, including User's and Producer's accuracy, and area uncertainty.
Include methods for accuracy visualization (plot and LaTeX tables)
Update data set names
Rename the data sets in ordes to avoid future overwriting of functions and data sets. "example_ts" replaced with "MOD13Q1.ts". Tthe data sets are now called:
MOD13Q1.MT.yearly.patterns Data: patterns time series MOD13Q1.patterns.list Data: patterns time series MOD13Q1.ts Data: An example of satellite time series MOD13Q1.ts.labels Data: Labels of the satellite time series in MOD13Q1.ts MOD13Q1.ts.list
Fix bug in twdtwApply wrong sign in 'by' argument
Fix bug in time index for twdtwApply-twdtwRaster
Include Fortran optimization
This version includes functions written in Fortran.
The S4 class 'twdtw' no longer exists.
New S4 classes: twdtwTimeSeries, twdtwMatches, and twdtwRaster.
plot methods for twdtwRaster object: 'maps', 'area', 'changes', and 'distance'.
plot methods for twdtwTimeSeries objects: ''patterns'' and ''timeseries''.
plot methods for twdtwMatches objects: ''paths'', ''matches'', ''alignments'', ''classification'', ''cost'', ''patterns'', and ''timeseries''.
createPattern function to create temporal patterns based on set of time series.
getTimeSeries extract time series from raster objects.
twdtwApply apply the TWDTW analysis for raster and time series objects.
'normalizeQuery' new normalization feature for TWDTW
'template.list' new dataset. List of template time series
arguments 'from' and 'to' in 'classifyIntervals' updated to include 'character' or 'Dates' in in the format 'yyyy-mm-dd'
Align query and template by name if names not null in 'twdtw' function
argument 'x' from function 'waveletSmoothing' is deprecated and is scheduled to be removed in the next version. Please use 'timeseries' instead.
argument 'template' from functions 'twdtw' and 'mtwdtw' is deprecated and is scheduled to be removed in the next version. Please use 'timeseries' instead.
argument 'normalized' is deprecated and is scheduled to be removed in the next version from all methods
'createTimeSequence' is deprecated. Use 'getModisTimeSequence' instead.
Fix function name. 'classfyIntervals' is deprecated. Use 'classifyIntervals' instead.
Fix plot intervals in plotClassify
replace range(x) for range(x, na.rm=TRUE) in all methods
Bug fixed in cost matrix indexing