International Conference on Computer and Knowledge Engineering

Home / 12th International Conference on Computer and Knowledge Engineering

Damage Detection After the Earthquake Using Sentinel-1 and 2 Images and Machine Learning Algorithms (Case Study: Sarpol-e Zahab Earthquake)

Authors :

Niloofar Alizadeh¹ Behnam Asghari Beirami² Mehdi Mokhtarzade³

1- Khajeh Nasir Toosi University of Technology 2- Khajeh Nasir Toosi University of Technology 3- Khajeh Nasir Toosi University of Technology

Keywords :

Change detection،Classification،Machine Learning،Damage،Earthquake

Abstract :

In remote sensing, by observing an area at different times, changes in the state of an object or phenomenon can be detected. Accurately identifying earthquake-affected areas can significantly aid in providing relief as soon as possible. This study proposes a simple hybrid method based on Sentinel-1 radar and Sentinel-2 optical images to detect damaged areas in Sarpol-e Zahab after the earthquake. This method employs a post-classification approach based on the decision fusion of optical and radar images to generate the change map in urban areas. Furthermore, this study employs Sentinel-1 radar images with the image ratio technique to detect the debris area accurately. The proposed method's change detection maps are visually compared to the European Space Agency's (ESA) produced damage map to validate the results. The final results reveal a good match between the detected damaged areas by the proposed method and the ESA product.