The research of underwater target recognition method based on deep learning

The deep learning is a popular research direction in machine learning field now. In this paper, the deep learning algorithms are used to recognize the underwater target radiated noises. The deep belief network (DBN) model and the stacked denoising autoencoder (SDAE) model are built respectively. Then the underwater acoustic simulated data of different types of targets as well as different states of one target and experimental data of different states of one target were recognized by these models, and the support vector machine (SVM), general regression neural network (GRNN) and probabilistic neural network (PNN) are selected as the comparison algorithms. The spectrum features extracted from the target radiated noises are used as the input data of the recognition models. The results show that the recognition accuracy of DBN and that of SDAE are all higher than that of the other three methods for all situations, which shows that the deep learning algorithms can effectively improve the underwater target recognition effect.