Analysis Of Supervised Classification Algorithms

Abstract Classification is the technique by which real world objectsland covers are identified within remotely sensed imagery. In supervised classification technique the location of land cover types should be known a priori. The areas of each land cover types are known as training sites. This classification is also termed as per-point or per-pixel classification. Accuracy and time complexity is observed for different levels of training dataset subjected to supervised classification algorithms. Finally different classifiers are compared for different levels of training set.