Small-sample large-scale regional landslide susceptibility evaluation method based on transfer learning

The invention discloses a small-sample large-scale regional landslide susceptibility evaluation method based on transfer learning, and the method comprises the following steps: S1, obtaining a historical landslide record of a research region, obtaining influence factor multi-source data, and unifying the spatial resolution, a projection coordinate system and a geographic coordinate system of the multi-source data; s2, acquiring an influence factor layer related to a landslide, extracting landslide influence factor information from the multi-source data through a GIS platform, and establishing a landslide susceptibility influence factor system; s3, on the basis of the landslide sample data set of the source domain, constructing a deep learning model, and pre-training the model; according to the method, the problem that the landslide susceptibility is predicted by using a machine learning method under the condition that the landslide sample size of a large-scale region is insufficient is solved; the investment