4,500 Seconds: Small Data Training Approaches for Deep UAV Audio Classification

Andrew P. Berg, Qian Zhang, Mia Y. Wang
5/21/2025
cs.SDcs.AIeess.AS

No official code found. Best implementation: AndrewPBerg/UAV_Classification (3 stars, MIT, last commit 2025-09-20).

Abstract

Unmanned aerial vehicle (UAV) usage is expected to surge in the coming decade, raising the need for heightened security measures to prevent airspace violations and security threats. This study investigates deep learning approaches to UAV classification focusing on the key issue of data scarcity. To investigate this we opted to train the models using a total of 4,500 seconds of audio samples, evenly distributed across a 9-class dataset. We leveraged parameter efficient fine-tuning (PEFT) and data augmentations to mitigate the data scarcity. This paper implements and compares the use of convolutional neural networks (CNNs) and attention-based transformers. Our results show that, CNNs outperform transformers by 1-2\% accuracy, while still being more computationally efficient. These early findings, however, point to potential in using transformers models; suggesting that with more data and further optimizations they could outperform CNNs. Future works aims to upscale the dataset to better understand the trade-offs between these approaches.

View on arXivView PDF

Code Implementations(1)

AndrewPBerg/UAV_ClassificationCommunity100%
MIT

Code behind "4,500 Seconds" and "15,500" Seconds. More to come

30Python, DockerfileSep 26, 20241 years agoMIT

Similar Papers

15,500 Seconds: Lean UAV Classification Using EfficientNet and Lightweight Fine-Tuning

87% match1 repo32025

From Ground to Air: Noise Robustness in Vision Transformers and CNNs for Event-Based Vehicle Classification with Potential UAV Applications

83% match2025

Object Detection in Thermal Images Using Deep Learning for Unmanned Aerial Vehicles

83% match2024

Predicting UAV Type: An Exploration of Sampling and Data Augmentation for Time Series Classification

82% match2024

Ensembling Pruned Attention Heads For Uncertainty-Aware Efficient Transformers

81% match2025

TakuNet: an Energy-Efficient CNN for Real-Time Inference on Embedded UAV systems in Emergency Response Scenarios

81% match1 repo542025

Cite this paper

@article{berg2025,
  title  = {4,500 Seconds: Small Data Training Approaches for Deep UAV Audio Classification},
  author = {Andrew P. Berg and Qian Zhang and Mia Y. Wang},
  year   = {2025},
  eprint = {2505.23782},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2505.23782v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2505.23782.svg)](https://codeofpaper.com/paper/2505.23782)

Discussion