Geological disaster identification method and device based on remote sensing image and DEM data

The invention discloses a geological disaster identification method and device based on remote sensing images and DEM data. The method comprises the following steps: constructing a geological disaster remote sensing image sample database; training a deep learning model by using the remote sensing image sample database; acquiring a debris flow disaster area and a non-debris flow disaster area by using the trained deep learning model according to a remote sensing image of a to-be-detected area; performing quantitative statistics on height and gradient aspects of each disaster in a non-debris flow disaster area by using DEM data; and acquiring a collapse area and a landslide area by using an integrated learning model according to the obtained statistical data. Therefore, by adopting the above scheme provided by the invention, high-efficiency, high-precision and high-robustness distinguishing of collapse, landslide and debris flow geological disasters is realized. The method solves the limitation of manual field