A Deep Learning Approach for Distant Infrasound Signals Classification

TL;DR

This paper introduces a deep learning framework that improves the classification of distant infrasound signals, achieving high accuracy on datasets involving explosions and seismic events.

Contribution

A novel spatiotemporal-based deep learning framework for infrasound classification, achieving 83.9% accuracy and outperforming existing methods.

Findings

The proposed framework achieves 83.9% classification accuracy on chemical explosion and seismic infrasound datasets.

The method outperforms eight other comparative classification approaches in experimental evaluations.

Incorporating spatiotemporal features enhances the accuracy of distant infrasound signal classification.

Abstract

Infrasound signal classification represents a critical challenge that demands immediate attention. Feature extraction stands as the core concept for enhancing classification accuracy in infrasound signal processing. However, existing feature extraction methodologies fail to meet the requirements for long-distance detection scenarios. To address these limitations, this study proposes a novel classification framework based on the spatiotemporal characteristics of infrasound signals. The proposed framework incorporates advanced signal processing techniques, signal enhancement algorithms, and deep learning architectures to achieve precise classification of infrasound signals. This paper designs three sets of comparative experiments, and the results demonstrate that the proposed method achieves a classification accuracy rate of 83.9% on chemical explosion and seismic infrasound datasets, outperforming eight other comparative classification methods. This substantiates the efficacy of the proposed approach.

Figures