<![CDATA[In the land cover classification process using the optical system remote sensing satellite data, there are problems in hilly areas where the lighting on the slopes facing or backward from the sun produces different spectral responses. In this study, we will analyze the effect of topographic correction on the Sun Canopy Sensor + C Correction (SCS + C) method on the accuracy of the classification results on the LANDSAT 8 surface reflectance image using Google Earth Engine (GEE). The results showed an increase in classification accuracy after topographic correction using the Support Vector Machine (SVM) method, Classification and Regression Tree (CRT), Random Forest (RF), and Minimum Distance (MD), respectively 4.45%, 3.33%, 2.23%, and 2.22%. The topographic correction applied to the Maximum Entropy (ME) classification methods failed to improve accuracy. It can be concluded that topographic correction can improve the accuracy of land cover classification results, especially in hilly areas]]>
Copyrights © 2020
