Random forest classifier for remote sensing classification

It is suggested that the random forest classifier performs equally well to SVMs in terms of classification accuracy and training time and the number of user‐defined parameters required byrandom forest classifiers is less than the number required for SVMs and easier to define.