sits (version 1.12.0)

sits_whittaker: Smooth the time series using Whittaker smoother

Description

The algorithm searches for an optimal polynomial describing the warping. Some authors consider the Whittaker smoother to be a good method for smoothing and gap filling for satellite image time series The degree of smoothing depends on smoothing factor lambda (usually from 0.5 to 10.0) Use lambda = 0.5 for very slight smoothing and lambda = 5.0 for strong smoothing

Usage

sits_whittaker(data.tb = NULL, lambda = 1, differences = 3,
  bands_suffix = "whit")

Arguments

data.tb

A tibble with time series data and metadata.

lambda

Smoothing factor to be applied (default 1.0).

differences

The order of differences of contiguous elements (default 3).

bands_suffix

Suffix to be appended to the smoothed filters (default "whit").

Value

A tibble with smoothed sits time series.

References

Francesco Vuolo, Wai-Tim Ng, Clement Atzberger, "Smoothing and gap-filling of high resolution multi-spectral timeseries: Example of Landsat data", Int Journal of Applied Earth Observation and Geoinformation, vol. 57, pg. 202-213, 2107.

Examples

# NOT RUN {
# Retrieve a time series with values of NDVI
data(point_ndvi)
# Filter the point using the whittaker smoother
point_ws.tb <- sits_whittaker (point_ndvi, lambda = 3.0)
# Plot the two points to see the smoothing effect
sits_plot(sits_merge(point_ndvi, point_ws.tb))
# }