Acoustic Classification of Maritime Vessels using Learnable Filterbanks

TL;DR

This paper introduces CATFISH, a deep learning model with learnable filterbanks that accurately classifies maritime vessels from acoustic data across diverse environments, achieving state-of-the-art performance.

Contribution

The paper presents a novel deep learning framework with a trainable spectral front-end for robust maritime vessel classification, outperforming existing benchmarks.

Findings

Achieved 96.63% test accuracy on VTUAD dataset.

Surpassed previous benchmark by over 12 percentage points.

Demonstrated effectiveness of learnable Gabor filterbanks in acoustic classification.

Abstract

Reliably monitoring and recognizing maritime vessels based on acoustic signatures is complicated by the variability of different recording scenarios. A robust classification framework must be able to generalize across diverse acoustic environments and variable source-sensor distances. To this end, we present a deep learning model with robust performance across different recording scenarios. Using a trainable spectral front-end and temporal feature encoder to learn a Gabor filterbank, the model can dynamically emphasize different frequency components. Trained on the VTUAD hydrophone recordings from the Strait of Georgia, our model, CATFISH, achieves a state-of-the-art 96.63 % percent test accuracy across varying source-sensor distances, surpassing the previous benchmark by over 12 percentage points. We present the model, justify our architectural choices, analyze the learned Gabor filters, and perform ablation studies on sensor data fusion and attention-based pooling.