Tool to support the selection of adequate satellite spatial resolution. Evaluates
how the spectral variability within a pixel change with the change in spatial resolution.
Usage
specVar(x, y)
Arguments
x
Object of class RasterLayer.
y
Spatial resolution (unit depends on the spatial projection).
Value
A list.
Details
Given a raster object (x), the function determines how degrading its spatial resolution
impacts our ability to perceive the complexity of the landscape. For the pixel resolution given by y,
The function resamples x and estimates the Mean Absolute Percentage Error (MAPE) for each pixel. The
MAPE is estimated as \(100 / n * sum(abs(O - A / O)\) where O are the original value in x,
A the aggregated value in the aggregated image and n the number of non-NA pixels in the original
image The output of the function consists of:
# NOT RUN { require(raster)
# read raster data r <- raster(system.file('extdata', '2013-07-16_ndvi.tif', package="rsMove"))
# apply function s.var <- specVar(r, 60)
# }