SAR ship target detection method based on balanced sample regression loss

The invention discloses an SAR ship target detection method based on balanced sample regression loss, which mainly solves the problem of low ship target detection performance of a trained network model due to difficult sample imbalance in an existing deep learning method. According to the implementation scheme, the method comprises the following steps: (1) obtaining ship data and dividing into training data and test data; 2) selecting a Faster- RCNN network as a training network model; 3) improving the original loss function of the training network to form a new total loss function; 4) sending training data into the network selected in the step 2), and training the network by using a new total loss function to obtain a finally trained network model; and 5) sending the test data to the trained network model to obtain a ship target detection result. According to the invention, the depth features of the ship target can be better extracted, the ship target detection performance is improved, and the method can be