A water surface target and underwater target classification method based on unsupervised domain adaptation

By constructing a UDA-ResNet classification model, utilizing modal domain beamforming and adversarial training, and combining simulation and experimental data, the problems of generalization and poor performance in water surface and underwater target classification under small sample conditions were solved, achieving effective classification under unsupervised conditions.

CN120724283BActive Publication Date: 2026-07-14BEIHANG UNIV +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIHANG UNIV
Filing Date
2025-06-13
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

Under small sample conditions, traditional deep learning methods have poor generalization and target classification performance, making it difficult to effectively distinguish between surface targets and underwater targets.

Method used

A domain-adaptive unsupervised classification method for surface and underwater targets is adopted. By constructing a UDA-ResNet classification model, modal domain beamforming and adversarial training are used, and simulation and experimental data are combined to extract domain-invariant features for classification.

Benefits of technology

It achieves effective classification of surface and underwater targets under small sample conditions, improving the model's generalization and classification performance without relying on a large amount of labeled experimental data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120724283B_ABST
    Figure CN120724283B_ABST
Patent Text Reader

Abstract

The application discloses a water surface target and underwater target classification method based on unsupervised domain adaptation, and belongs to the technical field of underwater acoustic target depth classification. The application solves the problems of poor generalization and target classification performance of traditional deep learning methods under small sample conditions. The classification model constructed by the application introduces sound field elevation angle structure simulation data to assist training, and extracts domain invariant features of simulation data and measured data through adversarial training, so that the classification performance degradation problem caused by the difference in feature distribution between domains is relieved. The application can realize effective classification of water surface targets and underwater targets under small sample conditions, and does not need to know the class labels of measured data, effectively improving the generalization and target classification performance of the model. The method can be applied to the classification of water surface targets and underwater targets.
Need to check novelty before this filing date? Find Prior Art