Satellite target recognition method based on deep learning

Aiming at the problem of satellite target recognition in space offensive and defense confrontation, a satellite target recognition method based on deep learning network is proposed. In this paper, Resnet101 network is constructed to extract features from ISAR images of different satellite targets by convolutional neural network, and finally realize satellite target recognition. The experimental results show that the average accuracy of the proposed method is 98.1% for the recognition of four different types of satellite targets, which can be well applied to the satellite target recognition in space environment, and has important practical value.