Underwater-art: Expanding information perspectives with text templates for underwater acoustic target recognition

Underwater acoustic target recognition is an intractable task due to the complex acoustic source characteristics and sound propagation patterns. Limited by insufficient data and narrow information perspective, recognition models based on deep learning seem far from satisfactory in practical underwater scenarios. Although underwater acoustic signals are severely influenced by distance, channel depth, or other factors, annotations of relevant information are often nonuniform, incomplete, and hard to use. In this work, the proposal is to implement underwater acoustic recognition based on templates made up of rich relevant information (UART). The templates are designed to integrate relevant information from different perspectives into descriptive natural language. UART adopts an audio-spectrogram-text trimodal contrastive learning framework, which endows UART with the ability to guide the learning of acoustic representations by descriptive natural language. These experiments reveal that UART has better recognition capability and generalization performance than traditional paradigms. Furthermore, the pretrained UART model could provide superior prior knowledge for the recognition model in the scenario without any auxiliary annotation.