Multispectral representation of Distributed Acoustic Sensing data: a framework for physically interpretable feature extraction and visualization

📅 2026-04-08
📈 Citations: 0
Influential: 0
📄 PDF

career value

212K/year
🤖 AI Summary
This study addresses the lack of standardized, physically interpretable visualization and feature representation methods for distributed acoustic sensing (DAS) data, which hinders large-scale analysis efficiency. To overcome this limitation, the work introduces multispectral imaging concepts into DAS for the first time, proposing a band-decomposition-based multispectral signal representation framework. By decomposing strain-rate signals into predefined frequency bands and generating corresponding band-energy images, the method constructs a spatiotemporal-spectral feature space with clear physical meaning that is amenable to automated processing. Integrated with unsupervised clustering and a ResNet-18 convolutional neural network, the approach achieves 97.3% accuracy in detecting cetacean vocalizations, significantly enhancing both the visual interpretability and automatic recognition performance of bioacoustic signals.

Technology Category

Application Category

📝 Abstract
Distributed Acoustic Sensing (DAS) enables continuous monitoring of dynamic strain along tens of kilometers of optical fiber, generating massive datasets whose interpretation and automated analysis remain challenging. DAS measurements often lack a standardized visual representation, and their physical interpretation depends strongly on acquisition conditions and signal processing choices. This work introduces a systematic framework for visualization and feature extraction of DAS data based on a multispectral signal representation. The approach decomposes strain-rate measurements into predefined frequency bands and computes band-limited energy images that describe the spatial and temporal distribution of acoustic energy across distinct spectral regimes. The framework is evaluated using DAS recordings containing Fin Whale (Balaenoptera physalus) and Blue Whale (Balaenoptera musculus) vocalizations. Three experiments are conducted to assess the approach: enhanced visualization of bioacoustic signals, unsupervised clustering of acoustic patterns, and supervised event detection using a convolutional neural network. Using multispectral composites as input, a ResNet-18 classifier achieves an accuracy of 97.3% in whale vocalization detection, demonstrating that the proposed representation captures biologically meaningful spectral structure and provides an effective feature space for automated analysis of DAS data.
Problem

Research questions and friction points this paper is trying to address.

Distributed Acoustic Sensing
data visualization
feature extraction
physical interpretability
multispectral representation
Innovation

Methods, ideas, or system contributions that make the work stand out.

Multispectral representation
Distributed Acoustic Sensing
Feature extraction
Physically interpretable
Whale vocalization detection
S
Sergio Morell-Monzó
Institut d’Investigació per a la Gestió Integrada de Zones Costaneres, Universitat Politècnica de València (IGIC-UPV), 46730 Gandia, Spain
D
Dídac Diego-Tortosa
Institut de Ciències del Mar (ICM), Consejo Superior de Investigaciones Científicas (CSIC), 08003 Barcelona, Spain
I
Isabel Pérez-Arjona
Institut d’Investigació per a la Gestió Integrada de Zones Costaneres, Universitat Politècnica de València (IGIC-UPV), 46730 Gandia, Spain
V
Víctor Espinosa
Institut d’Investigació per a la Gestió Integrada de Zones Costaneres, Universitat Politècnica de València (IGIC-UPV), 46730 Gandia, Spain