Bats, Spectrograms, and Deep Learning: Unmasking the Secrets of Mining Area Caves

  • Arthur Gonsales Instituto Tecnológico Vale
  • Vitor C. A. Santos Instituto Tecnológico Vale
  • Giulliana Appel Instituto Tecnológico Vale
  • Leonardo Trevelin Instituto Tecnológico Vale
  • Valeria Tavares Instituto Tecnológico Vale
  • Ronnie Alves Instituto Tecnológico Vale

Resumo


Detecting and classifying bat species in mining region caves is crucial for ore extraction activities, environmental impact reduction, and worker safety. This study focuses on using bat echolocation call data collected by VALE, a mining company, to develop a deep learning-based system for species recognition. By applying transfer learning on a pre-trained MobileNetV2 model, the system achieved an impressive 95.21% accuracy in classifying spectrograms of bat echolocation calls from three different species. This outperformed other models tested. Implementing this system would enhance VALE's cave inspection processes, ensuring worker safety and bat population preservation in mining regions.

Palavras-chave: deep learning, cave bats, spectrograms, classification

Referências

Arias-Aguilar, A. et al. (2018). Who’s calling? acoustic identification of brazilian bats. Mammal Research, 63:231–253.

Auler, A. S. and Pilo, L. B. (2014). Caves and mining in brazil: the dilemma of cave preservation within a mining context. In Hydrogeological and environmental investigations in karst systems, pages 487–496. Springer Berlin Heidelberg.

Falcao, F. et al. (2015). Unravelling the calls of discrete hunters: acoustic structure of echolocation calls of furipterid bats (chiroptera, furipteridae). Bioacoustics, 24(2):175–183.

Frazier, P. I. (2018). A tutorial on bayesian optimization. arXiv preprint arXiv:1807.02811.

Frick, W. F., Kingston, T., and Flanders, J. (2020). A review of the major threats and challenges to global bat conservation. Annals of the New York Academy of Sciences, 1469(1):5–25.

Gessinger, G. et al. (2019). Unusual echolocation behaviour of the common sword-nosed bat lonchorhina aurita: an adaptation to aerial insectivory in a phyllostomid bat? Royal Society open science, 6(7):182165.

Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.

Jacka, J. K. (2018). The anthropology of mining: the social and environmental impacts of resource extraction in the mineral age. Annual Review of Anthropology, 47:61–77.

Jones, K. E. et al. (2013). Indicator bats program: a system for the global acoustic monitoring of bats. In Biodiversity monitoring and conservation: bridging the gap between global commitment and local action, pages 211–247.

Kornblith, S., Shlens, J., and Le, Q. V. (2019). Do better imagenet models transfer better? In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2661–2671.

Liu, X. et al. (2023). Simple pooling front-ends for efficient audio classification. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE.

Mac Aodha, O. et al. (2018). Bat detective—deep learning tools for bat acoustic signal detection. PLoS computational biology, 14(3):e1005995.

Mantz, J. W. (2008). Improvisational economies: Coltan production in the eastern congo. Social Anthropology/Anthropologie Sociale, 16(1):34–50.

Michele, A., Colin, V., and Santika, D. D. (2019). Mobilenet convolutional neural networks and support vector machines for palmprint recognition. Procedia Computer Science, 157:110–117.

Ozer, I., Ozer, Z., and Findik, O. (2018). Noise robust sound event classification with convolutional neural network. Neurocomputing, 272:505–512.

Park, D. S. et al. (2019). Specaugment: A simple data augmentation method for automatic speech recognition. arXiv preprint arXiv:1904.08779.

Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.-C. (2018). Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4510–4520.

Smith, J. H. (2011). Tantalus in the digital age: Coltan ore, temporal dispossession, and “movement” in the eastern democratic republic of the congo. American Ethnologist, 38(1):17–35.

Sowmya, M., Balasubramanian, M., and Vaidehi, K. (2022). Classification of animals using mobilenet with svm classifier. In Computational Methods and Data Engineering: Proceedings of ICCMDE 2021, pages 347–358. Springer.

Tanveer, M. H. et al. (2021). Mel-spectrogram and deep cnn based representation learning from bio-sonar implementation on uavs. In 2021 International Conference on Computer, Control and Robotics (ICCCR), pages 220–224. IEEE.

Voigt, C. C. and Kingston, T. (2016). Bats in the Anthropocene: conservation of bats in a changing world. Springer Nature.

Walters, C., Collen, A., Lucas, T., Mroz, K., Sayer, C., and Jones, K. (2013). Challenges of using bioacoustics to globally monitor bats. In Bat Evolution, Ecology, and Conservation, page 479–99. Springer New York.

Yoh, N. et al. (2022). A machine learning framework to classify southeast asian echolocating bats. Ecological Indicators, 136:108696.

Zualkernan, I. et al. (2021). An aiot system for bat species classification. In 2020 IEEE International Conference on Internet of Things and Intelligence System (IoTaIS), pages 155–160. IEEE.
Publicado
25/09/2023
Como Citar

Selecione um Formato
GONSALES, Arthur; SANTOS, Vitor C. A.; APPEL, Giulliana; TREVELIN, Leonardo; TAVARES, Valeria; ALVES, Ronnie. Bats, Spectrograms, and Deep Learning: Unmasking the Secrets of Mining Area Caves. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 20. , 2023, Belo Horizonte/MG. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2023 . p. 1186-1194. ISSN 2763-9061. DOI: https://doi.org/10.5753/eniac.2023.234655.