Comparative Analysis of Deep Learning Architectures for Classifying Soy Seeds

• Vinicius Godoy Marques UNESP
• Lucas Correia Ribas UNESP
• Caetano Mazzoni Ranieri UNESP

DOI: https://doi.org/10.5753/eniac.2025.14087

Resumo

A qualidade das sementes de soja desempenha um papel crucial na produtividade agrícola. Estudos têm aplicado técnicas de aprendizado profundo para automatizar a classificação de sementes. No entanto, essas abordagens focam em classificações binárias ou não comparam múltiplas arquiteturas para a classificação multiclasse. Este trabalho propõe uma comparação de modelos para classificar sementes de soja em cinco categorias. A metodologia envolve a avaliação de diferentes modelos utilizando o conjunto de dados Soybean Seeds, seguido do ajuste fino do modelo com melhor desempenho. Os resultados mostram que o modelo InceptionV3 alcançou a maior acurácia, melhorando de 76,68% com aprendizado por transferência para 86% após ajuste fino.

Referências

Ayoub Shaikh, T., Rasool, T., and Rasheed Lone, F. (2022). Towards leveraging the role of machine learning and artificial intelligence in precision agriculture and smart farming. Computers and Electronics in Agriculture, 198:107119.

Bevers, N., Sikora, E. J., and Hardy, N. B. (2022). Soybean disease identification using original field images and transfer learning with convolutional neural networks. Computers and Electronics in Agriculture, 203:107449.

Chauhan, I., Kekre, S., Miglani, A., Kankar, P. K., and Ratnaparkhe, M. B. (2025). Cnn-based damage classification of soybean kernels using a high-magnification image dataset. Journal of Food Measurement and Characterization, 19(5):3471–3495.

Eryigit, R. and Tugrul, B. (2021). Performance of various deep-learning networks in the seed classification problem. Symmetry, 13(10).

He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep residual learning for image recognition.

Huang, Z., Wang, R., Cao, Y., Zheng, S., Teng, Y., Wang, F., Wang, L., and Du, J. (2022). Deep learning based soybean seed classification. Computers and Electronics in Agriculture, 202:107393.

Jogin, M., Mohana, Madhulika, M. S., Divya, G. D., Meghana, R. K., and Apoorva, S. (2018). Feature extraction using convolution neural networks (cnn) and deep learning. In 2018 3rd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), pages 2319–2323.

Lin, W., Fu, Y., Xu, P., Liu, S., Ma, D., Jiang, Z., Zang, S., Yao, H., and Su, Q. (2023). Soybean image dataset for classification. Data in Brief, 48:109300.

Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., and Xie, S. (2022). A convnet for the 2020s.

Pereira, G. M. L., Foleis, J. H., de Souza Brito, A., and Bertolini, D. (2024). A database for soybean seed classification. In 2024 37th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pages 1–6.

Saleem, M. H., Potgieter, J., and Arif, K. M. (2021). Automation in agriculture by machine and deep learning techniques: A review of recent developments. Precision Agriculture, 22:2053–2091.

Siamabele, B. (2021). Soybeans production, driving factors, and climate change perspectives. Journal for Creativity, Innovation and Social Entrepreneurship (JCISE), page 113.

Simonyan, K. and Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition.

Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., and Wojna, Z. (2015). Rethinking the inception architecture for computer vision.

Tan, M. and Le, Q. V. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. CoRR, abs/1905.11946.

Upadhyay, N. and Bhargava, A. (2025). Artificial intelligence in agriculture: applications, approaches, and adversities across pre-harvesting, harvesting, and post-harvesting phases. Iranian Journal of Computer Science.

Vogel, J. T. et al. (2021). Soybean yield formation physiology–a foundation for precision breeding based improvement. Frontiers in Plant Science, 12:719706.

Yang, S., Zheng, L., He, P., Wu, T., Sun, S., and Wang, M. (2021). High-throughput soybean seeds phenotyping with convolutional neural networks and transfer learning. Plant Methods, 17(1):50.

Yun, Y., Li, D., An, X., and Ma, Z. (2024). Research Progress on Seed Appearance Recognition for Major Crops, volume 9 of Smart Agriculture. Springer, Singapore.

Zhu, S., Zhang, J., Chao, M., Xu, X., Song, P., Zhang, J., and Huang, Z. (2020). A rapid and highly efficient method for the identification of soybean seed varieties: Hyperspectral images combined with transfer learning. Molecules, 25(1).