EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Comparing the Wavelet Scattering Transform and 1-D Convolutional Neural Networks for Drone Propeller Fault Classification

Erick Axel Martinez-Ríos, Riccardo Cespi, Stefano Di Gennaro, Martin Rogelio Bustamante-Bello

IEEE Access · 2026 · 14 · 97496-97525

DOI: 10.1109/access.2026.3706573 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Erick Axel Martinez-Ríos und 3 weiteren Autor:innen. Die Arbeit nennt 70 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Universitá degli Studi di L’Aquila under the International Agreement with the Instituto Tecnológico y de Estudios Superiores de Monterrey, Scholarship Awarded under the International Agreement between the Instituto Tecnológico y de Estudios Superiores de Monterrey and Universitá degli Studi di L’A….

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unmanned Aerial Vehicles (UAVs) are widely used in transportation, telecommunications, scientific mapping, and remote sensing. However, their diverse applications make them susceptible to propulsion system faults, particularly in the propellers. To address this concern, signal processing techniques combined with machine learning classifiers have been used to detect damage by analyzing quadrotor vibration data. Despite their effectiveness, these methods often depend on handcrafted features and require high computational resources, posing challenges for real-time embedded implementation. The wavelet scattering transform (WST) provides a robust and interpretable signal representation through cascaded convolution operations. While structurally similar to one-dimensional convolutional neural networks (1D CNNs), WST does not require training or large datasets. Compared to two-dimensional CNNs, 1D CNNs offer lower complexity and shallower architectures, making them more suitable for real-time quadrotor fault detection. However, limited work has explored the trade-offs between WST and 1D CNNs for this application. This study compares WST and 1D CNNs using vibration data in terms of classification performance and execution time. Support vector machines trained on WST coefficients achieve comparable accuracy and F1-score to 1D CNNs, while requiring significantly less computation time, with inference times under five milliseconds compared to over ten milliseconds for CNNs.

Förderung

  • Universitá degli Studi di L’Aquila under the International Agreement with the Instituto Tecnológico y de Estudios Superiores de Monterrey
  • Scholarship Awarded under the International Agreement between the Instituto Tecnológico y de Estudios Superiores de Monterrey and Universitá degli Studi di L’A…

Zitieren

APA

Martinez-Ríos, E. A., Cespi, R., Gennaro, S. D., & Bustamante-Bello, M. R. (2026). Comparing the Wavelet Scattering Transform and 1-D Convolutional Neural Networks for Drone Propeller Fault Classification. IEEE Access, 14, 97496-97525. https://doi.org/10.1109/access.2026.3706573

Vancouver

Martinez-Ríos EA, Cespi R, Gennaro SD, Bustamante-Bello MR. Comparing the Wavelet Scattering Transform and 1-D Convolutional Neural Networks for Drone Propeller Fault Classification. IEEE Access. 2026;14:97496-97525. doi:10.1109/access.2026.3706573

BibTeX
@article{MartinezRos2026,
  author = {Martinez-Ríos, Erick Axel and Cespi, Riccardo and Gennaro, Stefano Di and Bustamante-Bello, Martin Rogelio},
  title = {Comparing the Wavelet Scattering Transform and 1-D Convolutional Neural Networks for Drone Propeller Fault Classification},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {97496--97525},
  doi = {10.1109/access.2026.3706573}
}
Kostenlos starten →