Image similar water regimen mining and forecasting method based on rainfall spatial distribution

The invention discloses an image similar water regimen mining and forecasting method based on rainfall spatial distribution. The method comprises the following steps: S1, collecting rainfall runoff data of a drainage basin and dividing sessions; s2, interpolating observation station observation data, and making a rainfall spatial distribution sample image set; s3, preprocessing the sample image set, and dividing the sample image set into a training set and a test set; s4, improving a residual network model, improving a full connection layer into a global average pooling layer, transplanting pre-trained weights and parameters to the improved residual network model, constructing an input and output layer, calculating feature vectors of the rainfall spatial distribution map, calculating Euclidean distances of the feature vectors and taking the Euclidean distances as similarity, and constructing a complete grid rainfall similarity discrimination model; s5, searching a historical similar rainfall process by using