Galaxy Morphology Classification with Deep Neural Networks

• Antônio Chacarosque UFSC
• Alexandre L. Gonçalves UFSC
• Jim Lau UFSC
• Bernardo W. Borges UFSC
• Alison R. Panisson UFSC

DOI: https://doi.org/10.5753/eniac.2026.30581

Resumo

The large-scale analysis of astronomical data requires automated and reliable methods for interpreting visual information. In this work, we propose a deep learning-based approach for galaxy morphological classification using Convolutional Neural Networks (CNNs). Two architectures are evaluated: AlexNet and ResNet152. The models are trained and tested on a balanced subset of the Galaxy Zoo 2 dataset, considering nine morphological classes. Data augmentation and class balancing techniques are employed to address dataset limitations. Experimental results show that ResNet152 significantly outperforms AlexNet, achieving an accuracy of 91.96% against 76.48%. The results demonstrate the effectiveness of deep CNNs in extracting complex visual patterns from astronomical images, even under constrained datasets.

Referências

Amarù, S., Marelli, D., Ciocca, G., and Schettini, R. (2023). Dalib: A curated repository of libraries for data augmentation in computer vision. Journal of Imaging, 9(10):232.

Brownlee, J. (2018). Metrics to evaluate machine learning algorithms in python. [link]. Acesso em: 5 jun. 2025.

Buda, M., Maki, A., and Mazurowski, M. A. (2018). A systematic study of the class imbalance problem in convolutional neural networks. Neural Networks, 106:249–259.

Cao, J., Xu, T., Deng, Y., Deng, L., Yang, M., Liu, Z., and Zhou, W. (2024). Galaxy morphology classification based on convolutional vision transformer (cvt). Astronomy & Astrophysics, 683:A42.

Chen, L., Li, S., Bai, Q., Yang, J., Jiang, S., and Miao, Y. (2021). Review of image classification algorithms based on convolutional neural networks. Remote Sensing, 13(22):4712.

Chollet, F. (2017). Deep Learning with Python. Manning Publications.

Conselice, C. J., Wilkinson, A., Duncan, K., and Mortlock, A. (2016). The evolution of galaxy number density at z¡ 8 and its implications. The Astrophysical Journal, 830(2):83.

de Almeida, T. A. and Teixeira, R. R. P. (2024). Explorando a fronteira entre inteligência artificial e astronomia na educação. Informática na educação: teoria & prática, 27(1).

Dere, S., Fatima, M., Jagtap, R., Inamdar, U., and Shardoor, N. B. (2021). Anomaly detection in astronomical objects of galaxies using deep learning. In 2021 7th International Conference on Advanced Computing and Communication Systems (ICACCS), volume 1, pages 702–706. IEEE.

Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning. MIT Press.

He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778.

He, Y., Zhang, Y., Chen, S., and Hu, Y. (2023). Classification of galaxy morphology based on multi-channel deep residual networks. In 2023 IEEE 6th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), volume 6, pages 1648–1653. IEEE.

Hubble, E. P. (1926). Extra-galactic nebulae. Astrophysical Journal, 64, 321–369.

Kalvankar, S., Pandit, H., and Parwate, P. (2020). Galaxy morphology classification using efficientnet architectures. arXiv preprint arXiv:2008.13611.

Krizhevsky, A., Sutskever, I., and Hinton, G. E. (2017). Imagenet classification with deep convolutional neural networks. Communications of the ACM, 60(6):84–90.

Kumar, T., Brennan, R., Mileo, A., and Bendechache, M. (2024). Image data augmentation approaches: A comprehensive survey and future directions. IEEE Access.

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.

Misra, D., Mohanty, S. N., Agarwal, M., and Gupta, S. K. (2020). Convoluted cosmos: classifying galaxy images using deep learning. In Data Management, Analytics and Innovation: Proceedings of ICDMAI 2019, Volume 1, pages 569–579. Springer.

Rafsanjani, M. K., Bagherinezhad, H., and Batth, R. S. (2019). The classification of galaxy images using neural network algorithm. In 2019 International Conference on Computing, Power and Communication Technologies (GUCON), pages 127–131. IEEE.

Reza, M. (2021). Galaxy morphology classification using automated machine learning. Astronomy and Computing, 37:100492.

Sadasivam, T., Serrao, R. T., Bansod, A. P., Singh, T., and Bharathi, R. (2024). Celestlearn ai: Astronomical image generator and classifier. In 2024 Asia Pacific Conference on Innovation in Technology (APCIT), pages 1–6. IEEE.

Şenel, F. A. (2023). A hyperparameter optimization for galaxy classification. Computers, Materials & Continua, 74(2).