Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleAçık Erişim2020Comparison of tree-based classification algorithms in mapping burned forest areas; [Primerjava drevesnih algoritmov razvrščanja pri zajemu območij pogorelih gozdov]
Matci, Dilek Kucuk; Comert, Resul; Avdan, Ugur
In this study, we compared the performance of tree-based classification algorithms – Random Forest (RF), Rotation Forest (RotF), J48, The Alternating Decision Tree (ADTree), Forest by Penalising Attributes (Forest PA), Logical Analysis of Data Algorithm (LADTree) and Functional Trees (FT) – for mapping burned forest areas within the Mediterranean region in Turkey. Object-based image analysis (OBIA) was performed to pan-sharpened the Landsat 8 images. Four different burned areas, namely Kumluca, Adrasan, Anamur, and Alanya, were used as study areas. Kumluca, Anamur, and Alanya regions were used as training areas, and Adrasan region was used as the test area. Obtained results were evaluated with confusion matrix and statistically significant analysis. According to the results, FT and RotF produced more accurate results than other algorithms. Also, the results obtained with these algorithms are statistically significant.
- MakaleAçık Erişim2019OBJECT BASED BURNED AREA MAPPING WITH RANDOM FOREST ALGORITHM
Comert, Resul; Matcı, Dilek Kucuk; Avdan, Ugur
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. Remote sensing methods give advantages such as fast, easy-to-use and high accuracy for burned area mapping. Recent years machine learning algorithms have become more popular in satellite image classification, due to the effective solutions for the analysis of complex datasets which have a large number of variables. In this study, the success of object based random forest algorithm was investigated for burned forest area mapping. For this purpose, Object based image analysis (OBIA) was performed using Landsat 8 image of the Adrasan and Kumluca fires which occurred in 24 – 27 June 2016. The study consisted of five steps. In the first step, the multi-resolution image segmentation was performed for obtaining image objects from Landsat 8 spectral bands. In the second step, 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. Finally, 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.
- MakaleYalnız künye2019Mapping of shallow landslides with object-based image analysis from unmanned aerial vehicle data
Comert, Resul; Avdan, Ugur; Gorum, Tolga; Nefeslioglu, Hakan A.
The Black Sea Region of Turkey is one of the most landslide prone areas due to its high slope topography, heavy rainfall, and highly weathered hillslope material conditions. Preparation of landslide inventory maps is the first step in producing landslide susceptibility maps. Ground-based methods for mapping landslide occurrences are time-consuming and expensive. Additionally, landslide mapping based on satellite imageries and aerial photographs has some limitations, including climatic conditions, cost, and limited repetitive measurement capacity. Visual interpretation-based landslide mapping, which is based on satellite imageries and aerial photographs, is a time-consuming procedure that requires an experience-based expert opinion. Therefore, the data acquisition based on unmanned aerial vehicle (UAV) and landslide event inventory maps using an object-based classification approach can be superior to other methods in terms of speed and cost. In this study, we developed a semiautomatic model using object-based image analyses for rapid mapping of shallow landslides from the data obtained from UAVs after major landslide events in the Black Sea Region of Turkey. For this purpose, two test sites—Kurucasile (Bartin) and Cayeli (Rize)—were selected. Landslide mapping models were developed in the investigation sites, and the performance of the models was evaluated. The landslides' data obtained with the developed models were compared to the landslides' data produced by the experts. The comparison process revealed that landslides mapped by using UAV data have an accuracy rate higher than 86% according to the number of landslides and 83% according to the landslide area.