EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Identification and Counting of Pirapitinga Piaractus brachypomus Fingerlings Fish Using Machine Learning

Alene Santos Souza iD, Adriano Carvalho Costa iD, Heyde Francielle do Carmo França iD, Joel Jorge Nuvunga iD, Gidélia Araújo Ferreira de Melo iD, Lessandro do Carmo Lima iD, Vitória de Vasconcelos Kretschmer iD, Débora Ázara de Oliveira, Liege Dauny Horn iD, Isabel Rodrigues de Rezende iD, Marília Parreira Fernandes iD, Rafael Vilhena Reis Neto iD und 5 weitere

Animals · 17.10.2024 · 14 (20) · 2999

DOI: 10.3390/ani14202999 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Animals, online erschienen am 17.10.2024. Verfasst von Alene Santos Souza und 16 weiteren Autor:innen. Beteiligte Einrichtung: Department of Science Animal, Federal Institute of Education, Science and Technology of Goiás (IF Goiano), Campus Rio Verde, Goiana South Highway, Km…. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Center of Excellence in Agri-Food Systems and Nutrition—Eduardo Mondlane University—Mupato, Mozambique, Federal Institute of Goiás, Rio Verde Campus (IF Goiano), Coordination for the Improvement of Higher Education Personnel (CAPES).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Identifying and counting fish are crucial for managing stocking, harvesting, and marketing of farmed fish. Researchers have used convolutional networks for these tasks and explored various approaches to enhance network learning. Batch normalization is one technique that improves network stability and accuracy. This study aimed to evaluate machine learning for identifying and counting pirapitinga Piaractus brachypomus fry with different batch sizes. The researchers used one thousand photographic images of Pirapitinga fingerlings, labeled with bounding boxes. They trained the adapted convolutional network model with batch normalization layers added at the end of each convolution block. They set the training to one hundred and fifty epochs and tested batch sizes of 5, 10, and 20. Furthermore, they measured network performance using precision, recall, and mAP@0.5. Models with smaller batch sizes performed less effectively. The training with a batch size of 20 achieved the best performance, with a precision of 96.74%, recall of 95.48%, mAP@0.5 of 97.08%, and accuracy of 98%. This indicates that larger batch sizes improve accuracy in detecting and counting pirapitinga fry across different fish densities.

Schlagwörter

Stichwörter der Autor:innen

South American round fishaquacultureautomationdetection algorithmneural network

Autor:innen und Institutionen

  • Department of Science Animal, Federal Institute of Education, Science and Technology of Goiás (IF Goiano), Campus Rio Verde, Goiana South Highway, Km 01, Rio Verde 75901-970, GO, Brazil.
  • Center of Excellence in Agri-Food Systems and Nutrition, Eduardo Mondlane University, Julius Nyerere, n° 3453, Maputo P.O. Box 257, Mozambique.
  • Faculty of Science and Technology, Joaquim Chissano University, Grande Maputo Street, 88, Maputo P.O. Box 1110, Mozambique.
  • Department of Science Animal, State University Paulista Júlio de Mesquita Filho, Nelson Brihi Badur, 480, Registro 11900-000, SP, Brazil.
  • Department of Science Animal, Federal University of Lavras, Ignácio Valentin, Lavras 37200-900, MG, Brazil.
  • Department of Biological Sciences, College of Arts and Sciences, Washington State University, Pullman, WA 99163, USA.

ORCID-Profile: Alene Santos Souza, Adriano Carvalho Costa, Heyde Francielle do Carmo França, Joel Jorge Nuvunga, Gidélia Araújo Ferreira de Melo, Lessandro do Carmo Lima, Vitória de Vasconcelos Kretschmer, Liege Dauny Horn, Isabel Rodrigues de Rezende, Marília Parreira Fernandes, Rafael Vilhena Reis Neto, Rilke Tadeu Fonseca de Freitas, Rodrigo Fortunato de Oliveira, Pedro Henrique Viadanna, Cibele Silva Minafra

Förderung

  • Center of Excellence in Agri-Food Systems and Nutrition—Eduardo Mondlane University—Mupato, Mozambique
  • Federal Institute of Goiás, Rio Verde Campus (IF Goiano)
  • Coordination for the Improvement of Higher Education Personnel (CAPES)
  • National Council for Scientific and Technological Development (CNPq)
  • Goiás State Research Support Foundation (FAPEG)
  • Centre of Excellence in Exponential Agriculture (CEAGRE)

Zitieren

APA

Souza, A. S., Costa, A. C., França, H. F. D. C., Nuvunga, J. J., Ferreira de Melo, G. A., Lima, L. D. C., Kretschmer, V. V., de Oliveira, D. Á., Horn, L. D., Rezende, I. R., Fernandes, M. P., Reis Neto, R. V., Freitas, R. T. F., Oliveira, R. F., Viadanna, P. H., Vitorino, B. M., & Minafra, C. S. (2024). Identification and Counting of Pirapitinga Piaractus brachypomus Fingerlings Fish Using Machine Learning. Animals, 14(20), 2999. https://doi.org/10.3390/ani14202999

Vancouver

Souza AS, Costa AC, França HFDC, Nuvunga JJ, Ferreira de Melo GA, Lima LDC, et al. Identification and Counting of Pirapitinga Piaractus brachypomus Fingerlings Fish Using Machine Learning. Animals. 2024;14(20):2999. doi:10.3390/ani14202999

BibTeX
@article{Souza2024,
  author = {Souza, Alene Santos and Costa, Adriano Carvalho and França, Heyde Francielle do Carmo and Nuvunga, Joel Jorge and Ferreira de Melo, Gidélia Araújo and Lima, Lessandro do Carmo and Kretschmer, Vitória de Vasconcelos and de Oliveira, Débora Ázara and Horn, Liege Dauny and Rezende, Isabel Rodrigues de and Fernandes, Marília Parreira and Reis Neto, Rafael Vilhena and Freitas, Rilke Tadeu Fonseca de and Oliveira, Rodrigo Fortunato de and Viadanna, Pedro Henrique and Vitorino, Brenno Muller and Minafra, Cibele Silva},
  title = {Identification and Counting of Pirapitinga Piaractus brachypomus Fingerlings Fish Using Machine Learning},
  journal = {Animals},
  year = {2024},
  volume = {14},
  number = {20},
  pages = {2999},
  doi = {10.3390/ani14202999}
}
Kostenlos starten →