Remote sensing image ship target identification method based on ensemble learning and AIS data

The invention discloses a remote sensing image ship target identification method based on ensemble learning and AIS data. The method comprises the following steps: acquiring AIS data; performing feature extraction on the AIS data, obtaining ship appearance features and ship types, and generating a training data set; constructing an integrated learning model; dividing the training data set into at least two sub-data sets, and training a plurality of base classifier models by using one sub-data set; inputting the ship appearance features in the other sub-data set into a plurality of base classifier models, obtaining prediction classification results of the plurality of base classifier models, combining the prediction classification results with ship types to obtain a new data set, and training an integrated classifier model by using the new data set; obtaining the appearance characteristics of the ship target in the remote sensing image; and inputting the appearance features of the remote sensing image ship tar