CLASSIFICATION OF BURNED PEATLAND USING PROBABILISTIC NEURAL NETWORK ALGORITHM BASED ON HIGH TEMPORAL DATA

Abstract

Forest or land fires in Indonesia do not only occur in drylands but also in peatlands. Peatland fires are more dangerous and more difficult to overcome compared to non-peatland fires and the impacts of peatland fires are very harmful to society. One of the solutions in assessing forest and peatland fires is remote sensing technology. Satellite images obtained from remote sensing technology are usually classified for further analysis. The main objective of this study is to develop a classification model using Probabilistic Neural Network (PNN) to classify areas in peatland before, during, and after being burned on the satellite image Landsat 7 ETM +. Furthermore, the model is used to get the trajectory pattern of the burned area using the DBScan algorithm. The study area is Ogan Komering Ilir District, South Sumatera Province, image Landsat 7 ETM + taken from January 2015 – December 2015.