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

1. 1Generate beamformed audio maps using delay-and-sum beamforming on a multi-microphone array
2. 2Create binary supervision signals aligned with drone GPS telemetry
3. 3Train a U-Net model for spherical semantic segmentation to segment beamformed audio maps into regions of active sound presence
4. 4Evaluate the model's performance on low-SNR UAV detection tasks

This research benefits audio engineers and AI researchers working on UAV detection systems, as it provides a novel approach to acoustic source localization

💡 Dense beamformed energy maps and U-Net SELD can effectively detect UAVs in low-SNR environments

💡 U-Net model for 360-degree acoustic source localization improves low-SNR UAV detection