CNN Aplicada à Imagens de drone para Identificação de Famílias de Planta Daninha em pastagem
Abstract
New agricultural technologies such as drone and Convolutional Neural Networks (CNN) can favor organic and conventional crops. This work developed an algorithm based on CNN to identify, in drone images, weeds in pasture, Panelas/PE. Fabaceae had the best results for its morphological characteristics and for its species Crotalaria micans, toxic to cattle. The CNN obtained 100% accuracy, for 10 training epochs using internet, smartphone and drone images. Despite the good accuracy result, it was observed the occurrence of overfitting, caused by the low quantity of drone images and the frequency of rainfall in the state of Pernambuco, in 2022.
References
Aitkenhead, M. J. et al. Weed and crop discrimination using image analysis and artificial intelligence methods. Computers and electronics in Agriculture, v. 39, n. 3, p. 157-171, 2003.
Amaral, L. R. do et al. Aplicações de UAVs na Agricultura 4.0. Revista Ciência Agronômica, v. 51, n. spe, 2020.
Bharambe, R. Docker: empowering app development for developers. Disponível em: [link]. Acesso em: 29 jun. 2021.
Da Rosa, M. C. (2019). Redes neurais convolutivas aplicadas à detecção de ervas daninhas. Universidade Federal do Rio Grande do Sul. Escola de Engenharia. Programa de Pós-Graduação em Engenharia Elétrica.
Dias Filho, M. B. Degradação das pastagens: o que é como evitar. Brasília, DF: Embrapa, 2017. 19 p.
Esposito, M. et al. Drone and sensor technology for sustainable weed management: A review. Chemical and Biological Technologies in Agriculture, v. 8, n. 1, p. 1-11, 2021.
Jati, A., Georgiou, P. Neural predictive coding using convolutional neural networks toward unsupervised learning of speaker characteristics. IEEE/ACM Trans. Audio, Speech, Lang. Process., v. 27, n. 10, p. 1577–1589, 2019.
Liang, W-C.; Yang, Y-J.; Chao, C-M. Low-cost weed identification system using drones. In: 2019 Seventh International Symposium on Computing and Networking Workshops (CANDARW). IEEE, 2019. p. 260-263.
OpenDroneMap Authors ODM – A command line toolkit to generate maps, point clouds, 3D models and DEMs from drone, balloon or kite images. OpenDroneMap/ODM GitHub Page 2020; [link]. Acessado em 04 set. 2022.
Petrich, L. et al. Detection of Colchicum autumnale in drone images, using a machine-learning approach. Precision Agriculture, v. 21, n. 6, p. 1291-1303, 2020.
Rist, Y., Shendryk, I., Diakogiannis, F., Levick, S. Weed Mapping Using Very High Resolution Satellite Imagery and Fully Convolutional Neural Network. IGARSS, IEEE International Geoscience and Remote Sensing Symposium, p. 9784-9787, 2019.
Teixeira, T., Jayme, S., Arnal, G., Sônia, B., João, T., Neto, C., Vieira, L., Kleber, K., & Sampaio De Souza, X. (n.d.). Visão computacional aplicada na agricultura. Alice.Cnptia.Embrapa.Br. Retrieved August 16, 2022, from [link].
Tian, Y. Artificial Intelligence Image Recognition Method Based on Convolutional Neural Network Algorithm. IEEE Access, v. 8, p. 125731-125744, 2020.
Yang, J., Wang, Y., Chen, Y., Yu, J. Detection of Weeds Growing in Alfalfa Using Convolutional Neural Networks. Agronomy. v. 12, n. 6, p. 1459, 2022.
Zhang, W., Hansen, M., Volonakis, T., Smith, M., Smith. Broad-Leaf Weed Detection in Pasture. IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), p. 101-105, 2018.
Amaral, L. R. do et al. Aplicações de UAVs na Agricultura 4.0. Revista Ciência Agronômica, v. 51, n. spe, 2020.
Bharambe, R. Docker: empowering app development for developers. Disponível em: [link]. Acesso em: 29 jun. 2021.
Da Rosa, M. C. (2019). Redes neurais convolutivas aplicadas à detecção de ervas daninhas. Universidade Federal do Rio Grande do Sul. Escola de Engenharia. Programa de Pós-Graduação em Engenharia Elétrica.
Dias Filho, M. B. Degradação das pastagens: o que é como evitar. Brasília, DF: Embrapa, 2017. 19 p.
Esposito, M. et al. Drone and sensor technology for sustainable weed management: A review. Chemical and Biological Technologies in Agriculture, v. 8, n. 1, p. 1-11, 2021.
Jati, A., Georgiou, P. Neural predictive coding using convolutional neural networks toward unsupervised learning of speaker characteristics. IEEE/ACM Trans. Audio, Speech, Lang. Process., v. 27, n. 10, p. 1577–1589, 2019.
Liang, W-C.; Yang, Y-J.; Chao, C-M. Low-cost weed identification system using drones. In: 2019 Seventh International Symposium on Computing and Networking Workshops (CANDARW). IEEE, 2019. p. 260-263.
OpenDroneMap Authors ODM – A command line toolkit to generate maps, point clouds, 3D models and DEMs from drone, balloon or kite images. OpenDroneMap/ODM GitHub Page 2020; [link]. Acessado em 04 set. 2022.
Petrich, L. et al. Detection of Colchicum autumnale in drone images, using a machine-learning approach. Precision Agriculture, v. 21, n. 6, p. 1291-1303, 2020.
Rist, Y., Shendryk, I., Diakogiannis, F., Levick, S. Weed Mapping Using Very High Resolution Satellite Imagery and Fully Convolutional Neural Network. IGARSS, IEEE International Geoscience and Remote Sensing Symposium, p. 9784-9787, 2019.
Teixeira, T., Jayme, S., Arnal, G., Sônia, B., João, T., Neto, C., Vieira, L., Kleber, K., & Sampaio De Souza, X. (n.d.). Visão computacional aplicada na agricultura. Alice.Cnptia.Embrapa.Br. Retrieved August 16, 2022, from [link].
Tian, Y. Artificial Intelligence Image Recognition Method Based on Convolutional Neural Network Algorithm. IEEE Access, v. 8, p. 125731-125744, 2020.
Yang, J., Wang, Y., Chen, Y., Yu, J. Detection of Weeds Growing in Alfalfa Using Convolutional Neural Networks. Agronomy. v. 12, n. 6, p. 1459, 2022.
Zhang, W., Hansen, M., Volonakis, T., Smith, M., Smith. Broad-Leaf Weed Detection in Pasture. IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), p. 101-105, 2018.
Published
2023-08-06
How to Cite
VERÇOSA, João Pedro dos S.; MEDEIROS, Marcelo H. B.; SANTOS, Mário H. G.; SILVA, Suzanne S. M. L. C.; LIMA, Gustavo H. G.; TAVARES, Arthur C. F..
CNN Aplicada à Imagens de drone para Identificação de Famílias de Planta Daninha em pastagem. In: WORKSHOP ON COMPUTING APPLIED TO THE MANAGEMENT OF THE ENVIRONMENT AND NATURAL RESOURCES (WCAMA), 14. , 2023, João Pessoa/PB.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2023
.
p. 31-40.
ISSN 2595-6124.
DOI: https://doi.org/10.5753/wcama.2023.230680.
