# ggR

0th

Percentile

##### Plot RasterLayers in ggplot with greyscale

Plot single layer imagery in grey-scale. Can be used with any Raster* object.

##### Usage
ggR(img, layer = 1, maxpixels = 5e+05, alpha = 1, hue = 1,
sat = 0, stretch = "none", quantiles = c(0.02, 0.98),
coord_equal = TRUE, ggLayer = FALSE, ggObj = TRUE,
geom_raster = FALSE, forceCat = FALSE)
##### Arguments
img

raster

layer

Character or numeric. Layername or number. Can be more than one layer, in which case each layer is plotted in a subplot.

maxpixels

Integer. Maximal number of pixels to sample.

alpha

Numeric. Transparency (0-1).

hue

Numeric. Hue value for color calculation [0,1] (see \[grDevices]hsv). Change if you need anything else than greyscale. Only effective if sat > 0.

sat

Numeric. Saturation value for color calculation [0,1] (see \[grDevices]hsv). Change if you need anything else than greyscale.

stretch

Character. Either 'none', 'lin', 'hist', 'sqrt' or 'log' for no stretch, linear, histogram, square-root or logarithmic stretch.

quantiles

Numeric vector with two elements. Min and max quantiles to stretch to. Defaults to 2% stretch, i.e. c(0.02,0.98).

coord_equal

Logical. Force addition of coord_equal, i.e. aspect ratio of 1:1. Typically useful for remote sensing data (depending on your projection), hence it defaults to TRUE. Note however, that this does not apply if (ggLayer=FALSE).

ggLayer

Logical. Return only a ggplot layer which must be added to an existing ggplot. If FALSE s stand-alone ggplot will be returned.

ggObj

Logical. Return a stand-alone ggplot object (TRUE) or just the data.frame with values and colors

geom_raster

Logical. If FALSE uses annotation_raster (good to keep aestetic mappings free). If TRUE uses geom_raster (and aes(fill)). See Details.

forceCat

Logical. If TRUE the raster values will be forced to be categorical (will be converted to factor if needed).

##### Details

When img contains factor values and annotation=TRUE, the raster values will automatically be converted to numeric in order to proceed with the brightness calculation. <U+00E4>

The raster package provides a class lookup-table for categorical rasters (e.g. what you get if you run superClass in classification mode). If your raster has a lookup-table ggR will automatically treat it as categorical (see factor). However, the factor status of Raster objects is easily lost and the values are interpreted as numeric. In such cases you should make use of the forceCat = TRUE argument, which makes sure that ggplot2 uses a discrete scale, not a continuous one.

The geom_raster argument switches from the default use of annotation_raster to geom_raster. The difference between the two is that geom_raster performs a meaningful mapping from pixel values to fill colour, while annotation_raster is simply adding a picture to your plot. In practice this means that whenever you need a legend for your raster you should use geom_raster = TRUE. This also allows you to specify and modify the fill scale manually. The advantage of using annotation_raster (geom_raster = TRUE) is that you can still use the scale_fill* for another variable. For example you could add polygons and map a value to their fill colour. For more details on the theory behind aestetic mapping have a look at the ggplot2 manuals.

##### Value

 ggObj = TRUE: ggplot2 plot ggLayer = TRUE: ggplot2 layer to be combined with an existing ggplot2 ggObj = FALSE: data.frame in long format suitable for plotting with ggplot2, includes the pixel values and the calculated colors

• ggR
##### Examples
# NOT RUN {
library(ggplot2)
library(raster)
data(rlogo); data(lsat); data(srtm)

## Simple grey scale annotation
ggR(rlogo)

## With linear stretch contrast enhancement
ggR(rlogo, stretch = "lin", quantiles = c(0.1,0.9))

## ggplot with geom_raster instead of annotation_raster
## and default scale_fill*
ggR(rlogo, geom_raster = TRUE)

## with different scale
ggR(rlogo, geom_raster = TRUE) +
scale_fill_gradientn(name = "mojo", colours = rainbow(10)) +

## Plot multiple layers
# }
# NOT RUN {
ggR(lsat, 1:6, geom_raster=TRUE, stretch = "lin") +
scale_fill_gradientn(colors=grey.colors(100), guide = FALSE) +
theme(axis.text = element_text(size=5),
axis.text.y = element_text(angle=90),
axis.title=element_blank())

## Don't plot, just return a data.frame
df <- ggR(rlogo, ggObj = FALSE)
head(df, n = 3)

## Layermode (ggLayer=TRUE)
data <- data.frame(x = c(0, 0:100,100), y = c(0,sin(seq(0,2*pi,pi/50))*10+20, 0))
ggplot(data, aes(x, y)) +  ggR(rlogo, geom_raster= FALSE, ggLayer = TRUE) +
geom_polygon(aes(x, y), fill = "blue", alpha = 0.4) +
coord_equal(ylim=c(0,75))

## Categorical data
## In this case you probably want to use geom_raster=TRUE
## in order to perform aestetic mapping (i.e. a meaningful legend)
rc   <- raster(rlogo)
rc[] <- cut(rlogo[[1]][], seq(0,300, 50))
ggR(rc, geom_raster = TRUE)

## Legend cusomization etc. ...
ggR(rc, geom_raster = TRUE) + scale_fill_discrete(labels=paste("Class", 1:6))
# }
# NOT RUN {
## Creating a nicely looking DEM with hillshade background
terr <- terrain(srtm, c("slope", "aspect"))
hill <- hillShade(terr[["slope"]], terr[["aspect"]])
ggR(hill)

ggR(hill) +
ggR(srtm, geom_raster = TRUE, ggLayer = TRUE, alpha = 0.3) +
scale_fill_gradientn(colours = terrain.colors(100), name = "elevation")
# }

Documentation reproduced from package RStoolbox, version 0.2.6, License: GPL (>= 3)

### Community examples

Looks like there are no examples yet.