Acoustic Imaging for UAV Detection: Dense Beamformed Energy Maps and U-Net SELD

TL;DR

This paper presents a novel array-independent U-net model for 360-degree acoustic source localization, segmenting beamformed energy maps into active sound regions for robust drone detection and sound event localization.

Contribution

The work introduces a segmentation-based approach using U-net on beamformed maps, enabling array-independent, transferable drone detection and sound event localization.

Findings

U-net improves angular precision in drone localization.

Method generalizes across environments and microphone configurations.

Validated on drone recordings and DCASE benchmark for sound event detection.

Abstract

We introduce a U-net model for 360{\deg} acoustic source localization formulated as a spherical semantic segmentation task. Rather than regressing discrete direction-of-arrival (DoA) angles, our model segments beamformed audio maps (azimuth & elevation) into regions of active sound presence. Using delay-and-sum (DAS) beamforming on a custom 24-microphone array, we generate signals aligned with drone GPS telemetry to create binary supervision masks. A modified U-Net, trained on frequency-domain representations of these maps, learns to identify spatially distributed source regions while addressing class imbalance via the Tversky loss. Because the network operates on beamformed energy maps, the approach is inherently array-independent and can adapt to different microphone configurations and can be transferred to different microphone configurations with minimal adaptation. The segmentation outputs are post-processed by computing centroids over activated regions, enabling robust DoA estimates. Our dataset includes real-world open-field recordings of a DJI Air 3 drone, synchronized with 360{\deg} video and flight logs across multiple dates and locations. Experimental results show that U-net generalizes across environments, providing improved angular precision, offering a new paradigm for dense spatial audio understanding beyond traditional Sound Source Localization (SSL). We additionally validate the same beamforming-plus-segmentation formulation on the DCASE 2019 TAU Spatial Sound Events benchmark, showing that the approach generalizes beyond drone acoustics to multiclass Sound Event Localization and Detection (SELD) scenarios.