arealBias {nnDiag}R Documentation

Bias Assessment Using an Areal Approach

Description

Using the AOIs created from aoiMaker, this function makes an object that can be plotted to assess the areal bias of kNN classification. See McRoberts (2009) for a description of the plot.

Usage

arealBias(object, reference.set)

Arguments

object object of class "nnDaoi"
reference.set vector of reference set observations

Value

Returns an object of class "nnDarbias", which is a list containing the following:

knn.estimates vector of the mean values of each AOI
probability.estimates vector of the mean value of the reference points contained within each AOI
lower.confidence vector of the lower confidence interval values
upper.confidence vector of the upper confidence interval values

Author(s)

Brian Walters walte137@msu.edu

References

McRoberts, R.E. (2009) Diagnostic tools for nearest neighbors techniques when used with satellite imagery, Remote Sensing of Environment. 113, 489–499.

See Also

aoiMaker, bias

Examples

data(LuceVolume_subset)
LuceVolImg <- GDAL.open(system.file("data/LuceSubset_Volume.tif", package = "nnDiag"))
LuceMask <- GDAL.open(system.file("data/LuceSubset_Mask.tif", package = "nnDiag"))

## First create "nnDaoi" object
coords <- as.matrix(LuceVolume_subset[,10:11])
vol.aoi <- aoiMaker(coords, 3, LuceVolImg, LuceMask, aoi.size = 20,
min.points = 4)

ref.vol <- LuceVolume_subset$ref.volume

x <- arealBias(vol.aoi, ref.vol)
plot(x)

[Package nnDiag version 0.0-5 Index]