OBJECT BASED BURNED AREA MAPPING WITH RANDOM FOREST ALGORITHM

Abstract

It is very important to map the burned forest areas economically, quickly and with the high accuracy of issues such as damage assessment studies, fire risk analysis, and management of forest regeneration processes. Mapping burned areas with a fast and easy-to-use method and high accuracy will be a very useful tool for local forest management units. In this study, we developed the new approach, for mapping burned areas. In this regard we use the segmentation process to the image, then apply the random forest algorithm for obtaining the map of the burned areas. For this purpose, we use the Landsat 8 image of the Adrasan and Kumluca fires which occurred in 24 – 27 June 2016. The study consisted of four steps. After the multi-resolution image segmentation was performed on obtained image objects from Landsat 8 spectral bands, the image object metrics such as spectral index and layer values were calculated for all image objects. In the third step, a random forest classifier model was developed. Then, the developed model applied to the test site for classification of the burned area. The obtained results evaluated with confusion matrix based on the randomly sampled points. According to the results, we obtained 0.089 commission error (CE) with 0.014 omission error (OE). An overall accuracy was obtained as 0.99. The results show that this approach is very useful to be used to determine burned forest areas.