This classification can either be supervised or unsupervised.Supervised classification entails identifying a number of cluster centres using training areas. A basic example is an area of forest that has been marked as a training area for the class forest and an urban area that has been marked as urban.