A Hybrid Approach for Brazilian Coffee Scene Classification Using VGG16 and Image Meta-Features

• Guilherme Pança Franco UTFPR
• Guilherme Moreira Lima Furlaneto UTFPR
• José Luis Seixas Junior UEL
• Rafael Gomes Mantovani UTFPR

DOI: https://doi.org/10.5753/eniac.2026.30210

Resumo

Remote sensing image classification plays a crucial role in precision agriculture and land-use monitoring. However, identifying specific crops, such as coffee, remains challenging due to their high visual similarity to surrounding vegetation and the limitations of isolated classification paradigms. To overcome this, we propose and evaluate a hybrid architecture that combines deep semantic representations extracted by a pre-trained VGG16 network with 74 carefully selected handcrafted image meta-features. The proposed approach was evaluated on the Brazilian Coffee Scene Dataset across 30 independent runs, comparing its performance against traditional machine learning algorithms (e.g., Random Forest, SVM) and deep learning models (Custom CNN, VGG16). Experimental results demonstrated that the hybrid model achieved the highest average accuracy (0.891) and the greatest stability, outperforming all single-paradigm models. Statistical analysis via the Wilcoxon signed-rank test confirmed the significance of these improvements. Furthermore, feature importance and error analyses revealed that handcrafted descriptors provide genuine complementary evidence to deep representations, substantially reducing false negatives and improving robustness in coffee crop classification.

Referências

Aguiar, G. J., Mantovani, R. G., Mastelini, S. M., de Carvalho, A. C., Campos, G. F., and Junior, S. B. (2019). A meta-learning approach for selecting image segmentation algorithm. Pattern Recognition Letters, 128:480–487.

Belgiu, M. and Drǎguţ, L. (2016). Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing, 114:24–31.

Breiman, L. (2001). Random forests. Machine Learning, 45(1):5–32.

Cortes, C. and Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3):273–297.

Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 248–255.

Domingos, P. and Pazzani, M. (1997). On the optimality of the simple bayesian classifier under zero-one loss. Machine Learning, 29(2–3):103–130.

Haralick, R. M., Shanmugam, K., and Dinstein, I. (1973). Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3(6):610–621.

Hu, M.-K. (1962). Visual pattern recognition by moment invariants. IRE Transactions on Information Theory, 8(2):179–187.

Khatami, R., Mountrakis, G., and Stehman, S. V. (2016). A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes: General guidelines for practitioners and future research. Remote Sensing of Environment, 177:89–100.

LeCun, Y., Bengio, Y., and Hinton, G. (2015). Deep learning. Nature, 521(7553):436–444.

Mitchell, T. M. (1997). Machine Learning. McGraw-Hill.

Mountrakis, G., Im, J., and Ogole, C. (2011). Support vector machines in remote sensing: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 66(3):247–259.

Nogueira, K., Penatti, O. A. B., and dos Santos, J. A. (2017). Towards better exploiting convolutional neural networks for remote sensing scene classification. Pattern Recognition, 61:539–556.

Ojala, T., Pietikäinen, M., and Mäenpää, T. (2002). Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(7):971–987.

Pan, S. J. and Yang, Q. (2010). A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 22(10):1345–1359.

Penatti, O. A. B., Nogueira, K., and dos Santos, J. A. (2015). Do deep features generalize from everyday objects to remote sensing and aerial scenes domains? In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 44–51.

Ren, Q., Wu, Y., Zeng, Q., Yang, N., et al. (2025). A review of deep learning based agricultural remote sensing image segmentation. Journal of Agricultural Engineering.

Ribeiro, J., Mileski, M., Carvalho, L. F., Seixas Junior, J. L., and Mantovani, R. G. (2024). Image classification for precision agriculture: A coffee study. In Anais do XV Computer on the Beach (COTB). SBC.

Rizkita, A. K., Febrianto, A., Verdiana, M., Iqbal, A., and Algifari, M. H. (2026). Comparative analysis of deep learning architectures for coffee tree detection from aerial imagery. Journal of Applied Informatics and Computing, 10(2):1383–1390.

Rouse, J. W., Haas, R. H., Schell, J. A., and Deering, D. W. (1974). Monitoring vegetation systems in the great plains with erts. In Third Earth Resources Technology Satellite-1 Symposium, volume 1, pages 309–317.

Rumelhart, D. E., Hinton, G. E., and Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323(6088):533–536.

Simonyan, K. and Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition. In International Conference on Learning Representations (ICLR).

Wang, J., Wang, Y., Li, G., and Qi, Z. (2024). Integration of remote sensing and machine learning for precision agriculture: a comprehensive perspective on applications. Agronomy, 14(9):1975.

Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6):80–83.

Zhang, A., Lipton, Z. C., Li, M., and Smola, A. J. (2023). Dive into Deep Learning. Cambridge University Press. [link].