Acoustic classification of crabs with compact neural models
DOI:
https://doi.org/10.37537/rev.elektron.10.1.227.2026Keywords:
Biacoustics, Microcontrollers, Knowledge distillation, Embedded systems, Neural networksAbstract
This paper presents an acoustic classifier based on neural networks, designed for implementation in a portable system for the field detection and classification of signals emitted by Neohelice granulata and Cyrtograpsus angulatus. Using recordings acquired with a hydrophone in a controlled environment, pre-filtering and adaptive segmentation techniques are applied to extract 22 acoustic features used to train a low-dimensional neural network. Through knowledge distillation, different master model configurations, segmentation windows, and sampling schemes are analyzed to obtain an efficient model suitable for execution on low-cost microcontrollers. The final model is exported in TensorFlow Lite format, ready for integration into embedded platforms, achieving accuracies of approximately 80% with CPU and memory requirements compatible with this type of system.
Downloads
References
M. Minello, L. Calado, and F. C. Xavier, “Ecoacoustic indices in marine ecosystems: A review on recent developments, challenges, and future directions,” ICES Journal of Marine Science, vol. 78, no. 9, pp. 3066–3074, Oct. 2021. [Online]. Available: https://doi.org/10.1093/icesjms/fsab193
Ministerio de Ambiente y Desarrollo Sostenible de la Nación Argentina, “Ficha técnica de la Reserva de la Biosfera Mar Chiquita,” Buenos Aires, Argentina, 2023. [Online]. Available: https://www.argentina.gob.ar/sites/default/files/2023/02/fichas_web_07.pdf
Ambiente de Argentina, “Mar Chiquita,” s.f. [Online]. Available: https://ampargentina.org/areas/mar-chiquita/
(accessed Apr. 26, 2025).
M. P. Sal Moyano, M. Ceraulo, T. Luppi, M. A. Gavio, and G. Buscaino, “Anthropogenic and biological sound effects on the maternal care behavior of a key crab species,” Frontiers in Marine Science, vol. 10, 2023. [Online]. Available: https://www.frontiersin.org/articles/10.3389/fmars.2023.1050148
S. Kumar, D. Guruparan, P. Aaron, P. Telajan, K. Mahadevan, D. Davagandhi, and O. X. Yue, “Deep learning in computational biology: Advancements, challenges, and future outlook,” arXiv:2310.03086, 2023. [Online]. Available: https://arxiv.org/abs/2310.03086
D. Tuia et al., “Perspectives in machine learning for wildlife conservation,” Nature Communications, vol. 13, Art. no. 792, 2022.
A. Lamba, P. Cassey, R. Raja Segaran, and L. Koh, “Deep learning for environmental conservation,” Current Biology, vol. 29, pp. R977–R982, Oct. 2019.
M. Malik, U. Malik, and A. Malik, “Leveraging deep learning for accurate classification of Leptograpsus crabs based on morphological measurements,” in Intelligent Computing Systems, A. Safi, A. Martin-Gonzalez, C. Brito-Loeza, and V. Castañeda-Zeman, Eds. Cham, Switzerland: Springer Nature, 2025.
C. Wu, Z. Xie, K. Chen, C. Shi, Y. Ye, Y. Xin, R. Zarei, and G. Huang, “A part-based deep learning network for identifying individual crabs using abdomen images,” Frontiers in Marine Science, vol. 10, 2023. [Online]. Available: https://www.frontiersin.org/articles/10.3389/fmars.2023.1093542
Espressif Systems, ESP32 Series Datasheet, 2024. [Online]. Available: https://www.espressif.com/sites/default/files/documentation/esp32_datasheet_en.pdf
STMicroelectronics, “Artificial intelligence on STM32 microcontrollers,” 2024. [Online]. Available: https://www.st.com/en/embedded-software/x-cube-ai.html
STMicroelectronics, “STM32 microcontrollers,” 2024. [Online]. Available: https://www.st.com/stm32
V. V. Viswanathan, R. A. Chaitanya, R. Prasanna, P. C. Kakarla, V. S. P. J., and N. Mohan, “Implementation of tiny machine learning models on Arduino 33 BLE for gesture and speech recognition,” arXiv:2207.12866, 2022. [Online]. Available: https://arxiv.org/abs/2207.12866
Avisoft Bioacoustics, “UltrasoundGate 116H: USB-based ultrasound recording interface,” datasheet, 2023. [Online]. Available: https://avisoft.com/ultrasoundgate/116h/
Teledyne Marine / RESON, “TC4013 miniature reference hydrophone,” product leaflet / technical specifications, 2020. [Online]. Available: https://teramara.ca/sites/default/files/2022-01/reson-TC4013%20product%20leaflet.pdf
C. H. Lubba, S. S. Sethi, P. Knaute, S. R. Schultz, B. D. Fulcher, and N. S. Jones, “catch22: Canonical time-series characteristics selected through highly comparative time-series analysis,” Data Mining and Knowledge Discovery, vol. 33, no. 6, 2019.
M. Antonelli, “scrubDetection,” GitHub repository, 2025. [Online]. Available: https://github.com/maxanto/scrubDetection
(accessed Nov. 18, 2025).
J. Gou, B. Yu, S. J. Maybank, and D. Tao, “Knowledge distillation: A survey,” International Journal of Computer Vision, vol. 129, no. 6, pp. 1789–1819, 2021.
S. Bai, J. Z. Kolter, and V. Koltun, “An empirical evaluation of generic convolutional and recurrent networks for sequence modeling,” in Proc. Int. Conf. on Learning Representations (ICLR), 2018.
M. C. Cebedio, “Clasificación de cangrejos,” GitHub repository, 2025. [Online]. Available: https://github.com/cebedio/Clasificaci-n-de-cangrejos
M. C. Cebedio, L. A. Rabioglio, and L. De Micco, “Quantized generative autoencoder for audio spectrograms,” IEEE Embedded Systems Letters, 2025.
Google, “QKeras: Quantization extensions for Keras,” 2023. [Online]. Available: https://github.com/google/qkeras
TensorFlow, “TensorFlow Lite para microcontroladores,” 2025. [Online]. Available: https://www.tensorflow.org/lite/microcontrollers?hl=es-419
Downloads
Published
Issue
Section
License
Copyright (c) 2026 María Celeste Cebedio, Martín Lorusso, Leonardo Arnone, Lucas Rabioglio, Maximiliano Antonelli, Raul Lopresti, Luciana De Micco
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The authors who publish in this journal agree with terms established in the license Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
