logo

Convolutional Neural Network models as a tool for automating rock classification

10.24050/reia.v22i43.1813
Published 2025-01-01
Articles

Abstract

Classifying rocks into igneous, sedimentary, and metamorphic types at different scales is essential for studying a region. Technological advancements facilitate these observation and analysis processes, benefiting both the training of professionals and the accessibility of geological knowledge for all interested parties. As one of the machine learning tools, Convolutional Neural Networks (CNNs) excel in identifying and prioritizing object features in an image. Among the relevant models in this group is You Only Look Once (YOLO), which serves as a foundation for retraining more specialized models. This work aims to create retrained models based on YOLO to automate identifying and classifying different types of rocks.

 

Two YOLO architectures, YOLOv8n and YOLOv8x were retrained using rock sample images from online repositories and the catalog of the School of Geology of the Universidad Industrial de Santander. These images were verified and validated based on geological criteria. Throughout each training stage, changes were made to the image catalogs and training parameters. The validation of the trained models was performed using rock samples from the School of Geology that were not used in the training. To quantify the quality of the models, the calculation of the normalized confusion matrix was used based on the actual labels, thereby obtaining the percentage of correct identifications for each rock type proportionally.

 

A total of eleven retrained models were obtained during the different training stages. Two models achieved a higher generalization capacity, obtained values greater than 30% on the diagonal of the confusion matrices. Other two models reached values greater than 80% for identifying igneous rocks. For sedimentary rocks, one model achieved an accuracy of 93%, while for metamorphic rocks, another obtained an accuracy of 61%. These models can serve as a foundation for developing more complex models for specific projects. Even though they were not trained with outcrop images, it is possible to estimate the type of rock they belong to, highlighting the capability of these models to extract basic features.

Keywords

Categories

Authors

  • Sergio García-Arias

    Colombia

  • Francisco Alberto Velandia Patiño

    Colombia

References

Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., S.Corrado, G., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., … Zheng, X. (2015). TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. Zenodo. https://doi. org/10.5281/zenodo.12119782

Carolis, B. De, Ladogana, F., & Macchiarulo, N. (2020). YOLO TrashNet: Garbage Detection in Video Streams. 2020 IEEE Conference on Evolving and Adaptive Intelligent Systems (EAIS), 1–7. https://doi.org/10.1109/ EAIS48028.2020.9122693

Cheng, G., & Guo, W. (2017). Rock Images Classification by Using Deep Convolution Neural Network. Journal of Physics: Conference Series, 887, 012089. https://doi. org/10.1088/1742-6596/887/1/012089

Cracknell, M. J. (2014). Machine Learning for Geological Mapping: Algorithms and Applications [Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy, University of Tasmania]. https://doi.org/https://doi. org/10.25959/23236064.v1

Dramsch, J. S. (2020). 70 Years of Machine Learning in Geoscience in Review. En Advances in Geophysics (pp. 1–55). Elsevier Inc. https://doi.org/10.1016/ bs.agph.2020.08.002

García-Arias, S., Velandia, F., Sanabria Gómez, J. D., & Ulloque Ardila, M. T. (2024). Mapas Predictivos con Redes Neuronales a Partir de Propiedades Físicas de Las Rocas: Caso de Estudio en la Mesa de Los Santos (Santander). Revista EIA, 21(41). https://doi.org/10.24050/reia.v21i41.1731

Gómez, J., Sánchez, J., Ocampo, A., & Restrepo, J. W. (2013). Aplicación de Redes Neuronales en la Clasificación de Arcillas. Revista EIA, 9(17), 183–191. https:// revistas.eia.edu.co/index.php/reveia/article/view/459

Goyes-Peñafiel, P., & Hernandez-Rojas, A. (2021). Doble Evaluación de la Susceptibilidad por Movimientos en Masa Basada en Redes Neuronales Artificiales y Pesos de Evidencia. Boletín de Geología, 43(1). https://doi. org/10.18273/revbol.v43n1-2021009

Han, Z., Fang, Z., Li, Y., & Fu, B. (2023). A novel Dynahead-Yolo Neural Network for the Detection of Landslides with Variable Proportions Wsing Remote Sensing Images. Frontiers in Earth Science, 10. https://doi.org/10.3389/ feart.2022.1077153

Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning. Springer New York. https://doi.org/10.1007/978-0-387-84858-7

Hussain, M., Bird, J. J., & Faria, D. R. (2019). A Study on CNN Transfer Learning for Image Classification (pp. 191–202). https://doi.org/10.1007/978-3-319- 97982-3_16

Jocher, G., Chaurasia, A., & Qiu, J. (2023). Ultralytics YOLOv8. https://github.com/ ultralytics/ultralytics

Kawaguchi, K., Huang, J., & Kaelbling, L. P. (2019). Every Local Minimum Value Is the Global Minimum Value of Induced Model in Nonconvex Machine Learning. Neural Computation, 31(12), 2293–2323. https://doi.org/10.1162/ neco_a_01234

Li, J., Wu, Z., & Shen, Z. (2022). Open the Black Box – Visualising CNN to Understand Its Decisions on Road Network Performance Level. Promet, 34(4). https://doi. org/10.7307/ptt.v34i4.4037

Liu, X., Wang, H., Jing, H., Shao, A., & Wang, L. (2020). Research on Intelligent Identification of Rock Types Based on Faster R-CNN Method. IEEE Access, 8, 21804–21812. https://doi.org/10.1109/ACCESS.2020.2968515

MacKay, D. J. C. (2003). Information Theory, Inference, and Learning Algorithms. Cambridge University Press, Cambridge.

Młynarczuk, M., Górszczyk, A., & Ślipek, B. (2013). The Application of Pattern Recognition in the Automatic Classification of Microscopic Rock Images. Computers & Geosciences, 60, 126–133. https://doi.org/10.1016/j. cageo.2013.07.015

Patro, S., Jhariya, D. C., Sahu, M., Dewangan, P., & Dhekne, P. Y. (2022). Igneous Rock Cclassification Using Convolutional Neural Networks (CNN). IOP Conference Series: Earth and Environmental Science, 1032(1), 012045. https://doi. org/10.1088/1755-1315/1032/1/012045

Pratama, B. G., Qodri, M. F., & Sugarbo, O. (2023). Building YoloV4 Models for Identification of Rock Minerals in Thin Section. IOP Conference Series: Earth and Environmental Science, 1151(1), 012046. https://doi.org/10.1088/1755- 1315/1151/1/012046

Redmon, J., Divvala, S. K., Girshick, R. B., & Farhadi, A. (2015). You Only Look Once: Unified, Real-Time Object Detection. CoRR, abs/1506.02640. http://arxiv.org/ abs/1506.02640

Rodríguez-Guillen, R., Kern, J., & Urrea, C. (2024). Fast Rock Detection in Visually Contaminated Mining Environments Using Machine Learning and Deep Learning Techniques. Applied Sciences, 14(2), 731. https://doi.org/10.3390/ app14020731

Rohwer, R., Wynne-Jones, M., & Wysotzki, F. (1994). Neural Networks. En Machine Learning, Neural and Statistical Classification (pp. 84–106). Ellis Horwood.

Rojas, R. (1996). Neural Networks. Springer Berlin Heidelberg. https://doi. org/10.1007/978-3-642-61068-4

Saxena, N., Day-Stirrat, R. J., Hows, A., & Hofmann, R. (2021). Application of Deep Learning for Semantic Segmentation of Sandstone Thin Sections. Computers & Geosciences, 152, 104778. https://doi.org/10.1016/j.cageo.2021.104778

Shin, H.-C., Roth, H. R., Gao, M., Lu, L., Xu, Z., Nogues, I., Yao, J., Mollura, D., & Summers, R. M. (2016). Deep Convolutional Neural Networks for Computer- Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning. IEEE Transactions on Medical Imaging, 35(5), 1285–1298. https:// doi.org/10.1109/TMI.2016.2528162

Ślipek, B., & Młynarczuk, M. (2013). Application of Pattern Recognition Methods to Automatic Identification of Microscopic Images of Rocks Registered Under Different Polarization and Lighting Conditions. Geology, Geophysics & Environment, 39(4), 373. https://doi.org/10.7494/geol.2013.39.4.373

Srinivas Amiripalli, S., Nageshwara Rao, G., Behara, J., Sanjay Krishna, K., & Pavan Venkat Durga Ram, M. (2021). Mineral Rock Classification Using Convolutional Neural Network. https://doi.org/10.3233/APC210235

Szandała, T. (2023). Unlocking the Black Box of CNNs: Visualising the Decision- Making Process with PRISM. Information Sciences, 642, 119162. https://doi. org/10.1016/j.ins.2023.119162

Tarbuck, E. J., & Lutgens, F. K. (2005). Ciencias de la Tierra: una Introducción a la Geología Física (8a ed.). Pearson Educación S. A.

Van der Baan, M., & Jutten, C. (2000). Neural Networks in Geophysical Applications. GEOPHYSICS, 65(4), 1032–1047. https://doi.org/10.1190/1.1444797

Wada, K., mpitid, Buijs, M., N., Z. Ch., Kubovčík, Bc. M., Myczko, A., latentix, Zhu, L., Yamaguchi, N., Fujii, S., iamgd67, IlyaOvodov, Patel, A., Clauss, C., Kuroiwa, E., Iyengar, R., Shilin, S., Malygina, T., … Toft, H. (2021). wkentaro/labelme: v4.6.0. Zenodo. https://doi.org/10.5281/zenodo.5711226

Xin, M., & Wang, Y. (2019). Research on Image Classification Model Based on Deep Convolution Neural Network. EURASIP Journal on Image and Video Processing, 2019(1), 40. https://doi.org/10.1186/s13640-019-0417-8

Statistics

Abstract
0
PDF accesses
0

Dimensions

PlumX

Downloads

Download data is not yet available.