Farmland surface soil moisture inversion method based on CNN Sentinel multi-source data

The invention discloses a farmland surface soil moisture inversion method based on CNN Sentinel multi-source data, aiming at the influence of parameters such as vegetation types, density and the like on satellite data moisture inversion of Sentinel satellites, and high-precision inversion of farmland surface soil moisture can be realized. The method comprises the following steps: step 1) performing corresponding preprocessing on Sentinel satellite data before the Sentinel satellite data is used; 2) constructing a data set, and inputting characteristic parameters including a dual-polarization radar backscattering coefficient (sigma , sigma ), an altitude (H ), a local incident angle (LIA), polarization decomposition characteristics (H, A, alpha) and three vegetation indexes (NDVI, MSAVI, DVI); 3) dividing the prepared 154 samples into a training set and a test set according to a proportion of 3: 1, and taking the former as the training set of the model and the latter as the test set of the model; and 4) inputt