Predição de Velocidade de Ciclones Tropicais Utilizando Redes Neurais Convolutivas e Detecção de Movimento em Imagens
Resumo
Com o cenário de mudanças climáticas, o comportamento dos ciclones tropicais tem se tornado mais imprevisível. Atualmente, existem diversas abordagens que utilizam redes neurais convolutivas (CNN) com bons resultados de predição de velocidade para os ciclones. O objetivo desse trabalho é propor uma modificação do modelo de Rede Neural CNN-TC, adicionando um pré-processamento de detecção de movimento, obtido de imagens adjacentes temporalmente. Como resultado, o modelo proposto teve melhor acurácia, obtendo um valor de raiz do erro médio quadrático (RMSE) 7,55% menor do que o método original. Esse resultado nos motiva a desenvolver modelos mais complexos com informações de detecção de movimento.Referências
Chen, B., Chen, B.-F., and Lin, H.-T. (2018). Rotation-blended cnns on a new open dataset for tropical cyclone image-to-intensity regression. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 90–99. ACM.
Chen, B.-F., Chen, B., Lin, H.-T., and Elsberry, R. L. (2019). Estimating tropical cyclone intensity by satellite imagery utilizing convolutional neural networks. Weather and Forecasting, 34(2):447 – 465.
Chollet, F. et al. (2015). Keras. [link].
Deng, Z., Villarini, G., and Wang, Z. (2025). Climate change dominates over urbanization in tropical cyclone rainfall patterns. Communications Earth & Environment, 6(1):54.
Foundation, P. S. (2024). Python language reference, version 3.x. [link].
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1):1929–1958.
Zhang, C.-J., Chen, M.-S., Ma, L.-M., and Lu, X.-Q. (2025). Deep learning and wavelet transform combined with multi-channel satellite images for tropical cyclone intensity estimation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
Zhang, C.-J., Wang, X.-J., Ma, L.-M., and Lu, X.-Q. (2021). Tropical cyclone intensity classification and estimation using infrared satellite images with deep learning. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14:2070–2086.
Zhang, R., Liu, Q., and Hang, R. (2019). Tropical cyclone intensity estimation using two-branch convolutional neural network from infrared and water vapor images. IEEE Transactions on Geoscience and Remote Sensing, 58(1):586–597.
Chen, B.-F., Chen, B., Lin, H.-T., and Elsberry, R. L. (2019). Estimating tropical cyclone intensity by satellite imagery utilizing convolutional neural networks. Weather and Forecasting, 34(2):447 – 465.
Chollet, F. et al. (2015). Keras. [link].
Deng, Z., Villarini, G., and Wang, Z. (2025). Climate change dominates over urbanization in tropical cyclone rainfall patterns. Communications Earth & Environment, 6(1):54.
Foundation, P. S. (2024). Python language reference, version 3.x. [link].
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1):1929–1958.
Zhang, C.-J., Chen, M.-S., Ma, L.-M., and Lu, X.-Q. (2025). Deep learning and wavelet transform combined with multi-channel satellite images for tropical cyclone intensity estimation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
Zhang, C.-J., Wang, X.-J., Ma, L.-M., and Lu, X.-Q. (2021). Tropical cyclone intensity classification and estimation using infrared satellite images with deep learning. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14:2070–2086.
Zhang, R., Liu, Q., and Hang, R. (2019). Tropical cyclone intensity estimation using two-branch convolutional neural network from infrared and water vapor images. IEEE Transactions on Geoscience and Remote Sensing, 58(1):586–597.
Publicado
12/11/2025
Como Citar
SONEGO, Matheus Luis Machado; OLIVEIRA, Alessandro Bof de; BOF, Patricia; BARONE, Dante Augusto Couto.
Predição de Velocidade de Ciclones Tropicais Utilizando Redes Neurais Convolutivas e Detecção de Movimento em Imagens. In: ESCOLA REGIONAL DE APRENDIZADO DE MÁQUINA E INTELIGÊNCIA ARTIFICIAL DA REGIÃO SUL (ERAMIA-RS), 1. , 2025, Porto Alegre/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 312-315.
DOI: https://doi.org/10.5753/eramiars.2025.16675.