RStoolbox: validateMap – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

validateMap

Map accuracy assessment

Description

validate a map from a classification or regression model. This can be useful to update the accuracy assessment after filtering, e.g. for a minimum mapping unit.

Usage

validateMap(map, valData, responseCol, nSamples = 500,
  mode = "classification", classMapping = NULL)

Arguments

`map`	RasterLayer. The classified map.
`valData`	SpatialPolygonsDataFrame or SpatialPointsDataFrame with validation data.
`responseCol`	Character. Column containing the validation data in attribute table of `valData`.
`nSamples`	Integer. Number of pixels to sample for validation (only applies to polygons).
`mode`	Character. Either 'classification' or 'regression'.
`classMapping`	optional data.frame with columns `'class'` and `'classID'` defining the mapping from raster integers to class names.

Note

Polygons, which are smaller than the map resolution will only be considered if they overlap with a pixel center coordinate, otherwise they will be ignored.

Examples

## Not run: 
library(caret)
library(raster)

## Training data
data(lsat)
poly     <- readRDS(system.file("external/trainingPolygons.rds", package="RStoolbox"))

## Split training data in training and validation set (50%-50%)
splitIn   <- createDataPartition(poly$class, p = .5)[[1]]
train <- poly[splitIn,]
val   <- poly[-splitIn,]

## Classify (deliberately poorly)
sc <- superClass(lsat, trainData = train, responseCol = "class", nSamples = 50, model = "mlc")

## Polish map with majority filter

polishedMap <- focal(sc$map, matrix(1,3,3), fun = modal) 

## Validation
## Before filtering
val0 <- validateMap(sc$map, valData = val, responseCol = "class", 
                            classMapping = sc$classMapping)
## After filtering
val1 <- validateMap(polishedMap, valData = val, responseCol = "class",
                             classMapping = sc$classMapping)

## End(Not run)

RStoolbox

Tools for Remote Sensing Data Analysis

v0.2.6

GPL (>= 3)

Authors

Benjamin Leutner [cre, aut], Ned Horning [aut], Jakob Schwalb-Willmann [aut], Robert J. Hijmans [ctb]

Initial release