unsupervised classification with R

unsupervised classification with R

m

January 29, 2016

Here we see three simple ways to perform an unsupervised classification on a raster dataset in R. I will show these approaches, but first we need to load the relevant packages and the actual data. You could use the Landsat data used in the “Remote Sensing and GIS for Ecologists” book which can be downloaded here.

library("raster")  
library("cluster")
library("randomForest")

# loading the layerstack  
# here we use a subset of the Landsat dataset from "Remote Sensing and GIS for Ecologists" 
image <- stack("path/to/raster")
plotRGB(image, r=3,g=2,b=1,stretch="hist")

RGBimage

Now we will prepare the data for the classifications. First we convert the raster data in a matrix, then we remove the NA-values.

## returns the values of the raster dataset and write them in a matrix. 
v <- getValues(image)
i <- which(!is.na(v))
v <- na.omit(v)

The first classification method is the well-known k-means method. It separates n observations into  k clusters. Each observation belongs to the cluster with the nearest mean.

## kmeans classification 
E <- kmeans(v, 12, iter.max = 100, nstart = 10)
kmeans_raster <- raster(image)
kmeans_raster[i] <- E$cluster
plot(kmeans_raster)

Kmeans

The second classification method is called clara (Clustering for Large Applications). It work by clustering only a sample of the dataset and then assigns all object in the dataset to the clusters.

## clara classification 
clus <- clara(v,12,samples=500,metric="manhattan",pamLike=T)
clara_raster <- raster(image)
clara_raster[i] <- clus$clustering
plot(clara_raster)

clara

The third method uses a random Forest model to calculate proximity values. These values were clustered using k-means. The clusters are used to train another random Forest model for classification.

## unsupervised randomForest classification using kmeans
vx<-v[sample(nrow(v), 500),]
rf = randomForest(vx)
rf_prox <- randomForest(vx,ntree = 1000, proximity = TRUE)$proximity

E_rf <- kmeans(rf_prox, 12, iter.max = 100, nstart = 10)
rf <- randomForest(vx,as.factor(E_rf$cluster),ntree = 500)
rf_raster<- predict(image,rf)
plot(rf_raster)

randomForest

The three classifications are stacked into one layerstack and plotted for comparison.

class_stack <- stack(kmeans_raster,clara_raster,rf_raster)
names(class_stack) <- c("kmeans","clara","randomForest")

plot(class_stack)

Comparing the three classifications:

Looking at the different classifications we notice, that the kmeans and clara classifications have only minor differences.
The randomForest classification shows a different image.

 

want to read more about R and classifications? check out this book:

follow us and share it on:

you may also like:

Collaboration with Forest Ecology, Goettingen

Collaboration with Forest Ecology, Goettingen

A recent visit to the University of Göttingen provided the opportunity for collaboration talks with the group of Prof. Dr. Dominik Seidel, Faculty of Forest Sciences and Forest Ecology, Department for Spatial Structures and Digitization of Forests. Discussions focused...

From Kruger to Potchefstroom: Reconnecting with South African EAGLE

From Kruger to Potchefstroom: Reconnecting with South African EAGLE

After completing their internship in Kruger National Park, EAGLE students Sebastian and Clemens were not quite ready to leave South Africa behind. Instead of heading straight home, they reunited with their South African EAGLE friend, Charl Strydom, for a road trip...

The 6 Species of Remote Sensing Researchers

The 6 Species of Remote Sensing Researchers

A fun field guide of earth observation scientists at our EORC, a typology of 6 Species of Remote Sensing Researchers (we could not think of more yet ...) There’s a magical moment in every remote sensing get-together when six completely different personalities somehow...

Share This