SAR TARGET RECOGNITION METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

摘要

A SAR target recognition method and apparatus, a computer device, and a storage medium. The method comprises: obtaining SAR original image samples (S100); performing data enhancement on the SAR original image samples to generate an extended sample set (S200); building a basic residual neural network model, and adding residual control factors to the basic residual neural network model for optimization to build an optimized residual neural network model (S300); randomly extracting a certain number of pictures from the expanded sample set, and inputting the extracted pictures to the optimized residual neural network model for training to obtain a trained residual neural network model (S400); and inputting a SAR image to be detected to the trained network model for recognition, and outputting the recognition result (S500). The method effectively reduces over-fitting; moreover, the residual neural network model to which the residual control factors are added can improve the rate of convergence in the training process, thereby shortening the mode training time, and improving the efficiency and precision of target recognition.