EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Lightweight deep learning model for real-time acoustic bird pest detection on edge microcontrollers

Micheline Kazeneza, Anna Sergeevna Bosman, Destiny Kwabla Amenyedzi, Pacifique Niyongabire, Emmanuel Ndashimye, Damien Hanyurwimfura, Anthony Vodacek

PeerJ Computer Science · 10.06.2026 · 12 · e3925

DOI: 10.7717/peerj-cs.3925 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 10.06.2026. Verfasst von Micheline Kazeneza und 6 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Icipe-World Bank Financing Agreement, World Bank-Korea Trust Fund Agreement, for the PASET Regional Scholarship and Innovation, University of Rwanda’s African Centre of Excellence in the Internet of Things.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Agricultural pest management in resource-limited regions faces significant challenges from avian pests like Quelea species, which cause substantial grain crop losses. Yet current management relies mainly on labor-intensive manual monitoring, as the existing automated monitoring solutions remain prohibitively expensive. Acoustic monitoring on affordable microcontrollers offers a low-cost, viable solution for automating avian pest detection in smallholder farms across low-income countries. This article presents Enhanced MicroDSC (Microcontroller Depthwise Separable Convolution), an optimized depthwise separable convolution architecture for acoustic pest bird detection on low-cost microcontrollers. Audio data used to train the model were collected across three regions of Rwanda (Bugesera, Busogo, and Nyagatare) from grain crop farms cultivating maize, wheat, and rice. Recording sessions captured natural behavioral contexts across varied environmental conditions, seasons, and times of day to ensure representative real acoustic complexity and diversity. A dataset comprising 9,970 audio samples (1.5–5 s duration) representing 11 classes—eight pest species, two beneficial species, and one non-bird category—was constructed. The mel-frequency energy (MFE) feature was extracted and fed into the compared machine learning models. Rigorous evaluation of Enhanced MicroDSC across 10 independent training runs demonstrated the performance of 97.4 ± 2.3% accuracy, 97.5 ± 2.3% precision, 97.4 ± 2.3% recall, and 97.4 ± 1.5% F1-score. The model comprises 7,483 parameters, representing 97.6% and 99.9% reductions compared to standard DSC and traditional convolutional neural network (CNN) architectures, respectively, and 494 × fewer parameters than state-of-the-art models such as YAMNet (3.7M parameters). Practical feasibility was validated through deployment on a Seeed XIAO ESP32S3 microcontroller, demonstrating real-time inference capability with minimal resources. The proposed system offers an economically viable solution for automated pest detection in smallholder grain crop production systems, where avian pests pose critical threats to food security.

Förderung

  • Icipe-World Bank Financing Agreement
  • World Bank-Korea Trust Fund Agreement, for the PASET Regional Scholarship and Innovation
  • University of Rwanda’s African Centre of Excellence in the Internet of Things

Zitieren

APA

Kazeneza, M., Bosman, A. S., Amenyedzi, D. K., Niyongabire, P., Ndashimye, E., Hanyurwimfura, D., & Vodacek, A. (2026). Lightweight deep learning model for real-time acoustic bird pest detection on edge microcontrollers. PeerJ Computer Science, 12, e3925. https://doi.org/10.7717/peerj-cs.3925

Vancouver

Kazeneza M, Bosman AS, Amenyedzi DK, Niyongabire P, Ndashimye E, Hanyurwimfura D, et al. Lightweight deep learning model for real-time acoustic bird pest detection on edge microcontrollers. PeerJ Computer Science. 2026;12:e3925. doi:10.7717/peerj-cs.3925

BibTeX
@article{Kazeneza2026,
  author = {Kazeneza, Micheline and Bosman, Anna Sergeevna and Amenyedzi, Destiny Kwabla and Niyongabire, Pacifique and Ndashimye, Emmanuel and Hanyurwimfura, Damien and Vodacek, Anthony},
  title = {Lightweight deep learning model for real-time acoustic bird pest detection on edge microcontrollers},
  journal = {PeerJ Computer Science},
  year = {2026},
  volume = {12},
  pages = {e3925},
  doi = {10.7717/peerj-cs.3925}
}
Kostenlos starten →