Deep learning for segmentation and classification of rock grains in the aggregates industry

  • Ricardo Ramos Nunes Centro Universitário Facens
  • Renato Moraes Silva USP

Resumo


This paper describes the application of a convolutional neural network (CNN) to detect rock grains in aggregates that exceed a specified size. In civil construction, the quality of aggregates is crucial and generally assessed by granulometry, with traditional sieving methods being time-consuming and susceptible to human error. This study proposes using machine learning to measure grain size continuously during the production process. Using CNN with the U-net architecture, the models were trained to evaluate grain size in a simulated condition that reproduces the environment of a conveyor belt. The results indicate that the developed models have strong generalization capabilities and can effectively identify contamination by rock grains that exceed the permitted size.

Palavras-chave: U-Net architecture, semantic segmentation, rock grain segmentation

Referências

Alom, M. Z., Hasan, M., Yakopcic, C., Taha, T. M., and Asari, V. K. (2018). Recurrent residual convolutional neural network based on u-net (r2u-net) for medical image segmentation. arXiv preprint arXiv:1802.06955.

Associação Brasileira de Normas Técnicas (2005). NBR 7211: Agregados para concreto - Especificação. Rio de Janeiro. Acesso em: 27 abril 2024.

Bamford, T., Esmaeili, K., and Schoellig, A. P. (2021). A deep learning approach for rock fragmentation analysis. International Journal of Rock Mechanics and Mining Sciences, 145:104839.

Canny, J. (1986). A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 8(6):679–698.

Chen, L., Zhou, X., Wang, M., Qiu, J., Cao, M., and Mao, K. (2019). ARU-Net: Research and application for wrist reference bone segmentation. IEEE Access, 7:166930–166938.

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

Haralick, R. M., Sternberg, S. R., and Zhuang, X. (1987). Image analysis using mathematical morphology. IEEE Transactions on Pattern Analysis and Machine Intelligence, 9(4):532–550.

Hassan, S. M., Laban, N., Abo Khashaba, S. M., El-Shibiny, N. H., Bashir, B., Azer, M. K., Drüppel, K., and Keshk, H. M. (2024). Semantic segmentation of some rock forming mineral thin sections using deep learning algorithms: A case study from the nikeiba area, south eastern desert, egypt. Remote Sensing, 16(13).

LeCun, Y., Jackel, L., Bottou, L., Brunot, A., Cortes, C., Denker, J., Drucker, H., Guyon, I., Muller, U., Sackinger, E., et al. (1995). Comparison of learning algorithms for hand-written digit recognition. In International conference on artificial neural networks, volume 60, pages 53–60. Perth, Australia.

Leiva, C., Acuña, C., and Castillo, D. (2021). Development and validation of an online analyzer for particle size distribution in conveyor belts. Minerals, 11(6):581.

Maitre, J., Bouchard, K., and Bédard, L. P. (2019). Mineral grains recognition using computer vision and machine learning. Computers & Geosciences, 130:84–93.

MT Expo (2022). Demanda global por areia e agregados tende a subir 45% até 2060. Accessed: 2024-05-17.

Oktay, O., Schlemper, J., Le Folgoc, L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y., Kainz, B., et al. (2018). Attention u-net: Learning where to look for the pancreas. In Medical Imaging with Deep Learning, pages 63–72.

Přikryl, R. (2021). Geomaterials as construction aggregates: a state-of-the-art. Bulletin of Engineering Geology and the Environment, 80(12):8831–8845.

Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18, pages 234–241. Springer.

Shi, P., Duan, M., Yang, L., Feng, W., Ding, L., and Jiang, L. (2022). An improved u-net image segmentation method and its application for metallic grain size statistics. Materials, 15(13):4417.

Silva, D. d. A. and Geyer, A. L. B. (2018). Análise e classificação da forma do agregado graúdo britado para concreto. Revista Científica Multidisciplinar Núcleo do Conhecimento ISSN, 2448:0959.

Wang, F. and Zai, Y. (2023). Image segmentation and flow prediction of digital rock with u-net network. Advances in Water Resources, 172:104384.

Wang, Y., He, Z., Xie, P., Yang, C., Zhang, Y., Li, F., Chen, X., Lu, K., Li, T., Zhou, J., and Zuo, K. (2021a). Attention recurrent residual convolutional network for multi-organ segmentation in medical imaging. Security and Communication Networks, 2021:1–9.

Wang, Y., Ma, G., Mei, J., Zou, Y., Zhang, D., Zhou, W., and Cao, X. (2021b). Machine learning reveals the influences of grain morphology on grain crushing strength. Acta Geotechnica, 16(11):3617–3630.

Weir Motion Metrics (2024). Solução da empresa weir para análise de partículas em tempo real. [link]. Acesso em: 27 abril 2024.

Yang, D., Wang, X., Zhang, H., yu Yin, Z., Su, D., and Xu, J. (2021). A mask r-cnn based particle identification for quantitative shape evaluation of granular materials. Powder Technology, 392:296–305.

Zhong, X., Deetman, S., Tukker, A., and Behrens, P. (2022). Increasing material efficiencies of buildings to address the global sand crisis. Nature Sustainability, 5(5):389–392.
Publicado
17/11/2024
NUNES, Ricardo Ramos; SILVA, Renato Moraes. Deep learning for segmentation and classification of rock grains in the aggregates industry. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 21. , 2024, Belém/PA. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2024 . p. 144-155. ISSN 2763-9061. DOI: https://doi.org/10.5753/eniac.2024.245219.