Deep learning based methods have achieved state-of-the-art results on the task of ship type classification. However, most existing ship type classification algorithms take time–frequency (TF) features as input, the underlying discriminative information of these features has not been explored thoroughly. This paper proposes a novel feature optimization method which is designed to minimize an objective function aimed at increasing inter-class and reducing intra-class feature distance for ship type classification. The objective function we design is able to learn a center for each class and make samples from the same class closer to the corresponding center. This ensures that the features maximize underlying discriminative information involved in the data, particularly for some targets that usually confused by the conventional manual designed feature. Results on the dataset from a real environment show that the proposed feature optimization approach outperforms traditional TF features.

Underwater target classification is an important research direction in sonar signal processing. There are two core parts in this area: feature extraction and classifier design. In this paper, a classifier based on time delay neural network (TDNN) was proposed, which has great advantages of modeling the temporal dynamics and representing the complex nonlinear relationship. Filter bank (FBANK) is used to extract features containing spectrum information as the input of the network. The proposed TDNN based classifier is evaluated on simulated data and real experimental data separately. The experimental results showed that the proposed method has higher classification accuracy than traditional methods such as support vector machine (SVM). Additionally, more details are compared in different feature dimensions and different SNRs. The method we proposed is also verified in real environmental data. More than 90% accuracy is achieved in three-classification experiment in real environment.