In recent years the popularity of tree based ensemble methods for land cover classification has increased significantly. Using WorldView-2 image data, we evaluate the potential of the oblique Random Forest algorithm (oRF) to classify a highly heterogeneous protected area. In contrast to the Random Forest (RF) algorithm, the oRF algorithm builds multivariate trees by learning the optimal split using a supervised model. In this study, the oRF binary algorithm is adapted to a multiclass land cover and land use application using both the ‘one-against-one’ and ‘one-against-all’ combination approaches. Results show that the oRF algorithms are capable of achieving high classification accuracies (> 80%). However, there was no statistical difference in classification accuracies obtained by the oRF algorithms and the more popular RF algorithm. For all the algorithms, user and producer accuracies greater than 80% were recorded for most of the classes. Both the RF and oRF algorithms poorly classified the indigenous forest class as indicated by the low user and producer accuracies. Finally, the results from this study advocate and support the utility of the oRF algorithm for land cover and land use mapping of protected areas using WorldView-2 image data.

BASSA Zaakirah;  BOB Urmilla;  SZANTOI Zoltan;  ISMAIL Riyad; 

2016-11-10

SPIE-SOC PHOTO-OPTICAL INSTRUMENTATION ENGINEERS

JRC100522

1931-3195,   

http://remotesensing.spiedigitallibrary.org/article.aspx?articleid=2503738,    https://publications.jrc.ec.europa.eu/repository/handle/JRC100522,   

10.1117/1.JRS.10.015017,