Computers and Concrete

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Prediction of concrete slump by RGB-D image feature fusion

  • Huansen Chen (College of Mechanical Engineering and Automation, Huaqiao University) ;
  • Jianhong Yang (College of Mechanical Engineering and Automation, Huaqiao University) ;
  • Huaiying Fang (College of Mechanical Engineering and Automation, Huaqiao University) ;
  • Shaojie Wu (College of Mechanical Engineering and Automation, Huaqiao University) ;
  • Bohong Lin (College of Mechanical Engineering and Automation, Huaqiao University)
  • Received : 2023.08.21
  • Accepted : 2024.02.26
  • Published : 2024.11.25
⟨ Previous Next ⟩

Abstract

Slump is an important index for concrete fluidity, which has a direct guiding effect on construction. In recent years, using RGB images for evaluating slump has been confirmed by scholars. Based on previous studies, this paper investigates the superiority of RGB-D image data over RGB image data in predicting slump of concrete and proposes three RGB-D fusion models: The early-stage-fusion model performs feature fusion in the data input stage, while the fully-connected-layer-fusion model performs feature fusion in the classification layer and the middle-stage-fusion model performs feature fusion after each residual block. In the classification of slump 120 mm, 150 mm and 200 mm, the Precision, Recall and F1-score are used to evaluate the model's ability to classify a single class, and the Accuracy, Macro-F1, Kappa and MCC are used to evaluate the model's performance. The experimental results showed that compared with the model using only RGB images, the fusion model achieve better performance, indicating that RGB-D image data can better evaluate concrete slump.

Keywords

Acknowledgement

This research is financially supported by the Major Program of Industry and University Cooperation of Fujian Province (2024H6010) and Quanzhou Introduces High-level Talent Team Program (2023CT003).

References

  1. Andayani, R. and Madenda, S. (2016), "Concrete slump classification using GLCM feature extraction", IOP Conf. Ser.: Mater. Sci. Eng., 131(1), 012011. https://doi.org/10.1088/1757-899X/131/1/01201.
  2. Baudez, J.C., Chabot, F. and Coussot, P. (2002), "Rheological interpretation of the slump test", Appl. Rheol., 12(3), 133-141. https://doi.org/10.1515/arh-2002-0008.
  3. Chandwani, V., Agrawal, V. and Nagar, R. (2014), "Modeling and analysis of concrete slump using hybrid artificial neural networks", Int. J. Civil Struct. Constr. Archit. Eng., 8(9), 933-940.
  4. Cihan, M.T. (2019), "Prediction of concrete compressive strength and slump by machine learning methods", Adv. Civil Eng., 2019, 1-11. https://doi.org/10.1155/2019/3069046.
  5. Ding, Z. and An, X. (2018), "Deep learning approach for estimating workability of self-compacting concrete from mixing image sequences", Adv. Mater. Sci. Eng., 2018, 1-16. https://doi.org/10.1155/2018/6387930.
  6. Ding, Z., An, X. (2017), "A method for real-time moisture estimation based on self-compacting concrete workability detected during the mixing process", Constr. Build. Mater., 139, 123-131. https://doi.org/10.1016/j.conbuildmat.2017.02.047.
  7. Emad, W., Mohammed, A.S., Kurda, R., Ghafor, K., Cavaleri, L., Qaidi, S.M., Hassan, A.M.T. and Asteris, P.G. (2022), "Prediction of concrete materials compressive strength using surrogate models", Struct., 46, 1243-1267. https://doi.org/10.1016/j.istruc.2022.11.002.
  8. Ferraris, C.F. and de Larrard, F. (1998), "Modified slump test to measure rheological parameters of fresh concrete", Cement Concrete Aggreg., 20(2), 241-247. https://doi.org/10.1520/CCA10417J.
  9. He, K., Zhang, X., Ren, S. and Sun, J. (2016), "Deep residual learning for image recognition", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, June.
  10. Kaloop, M.R., Samui, P., Shafeek, M. and Hu, J.W. (2020), "Estimating slump flow and compressive strength of self-compacting concrete using emotional neural networks", Appl. Sci., 10(23), 8543. https://doi.org/10.3390/app10238543.
  11. Kumar, R. and Mai, H.V.T. (2022), "Prediction and sensitivity analysis of self-compacting concrete slump flow by random forest algorithm", J. Sci. Transp. Technol., 2(1), 32-43. https://doi.org/10.58845/jstt.utt.2022.en.2.1.32-43.
  12. Li, S. and An, X. (2014), "Method for estimating workability of self-compacting concrete using mixing process images", Comput. Concrete, 13(6), 781-798. http://doi.org/10.12989/cac.2014.13.6.781.
  13. Liao, S., Li, G., Li, J., Jiang, D., Jiang, G., Sun, Y., Bo, T., Haoyi, Z. and Chen, D. (2021), "Multi-object intergroup gesture recognition combined with fusion feature and KNN algorithm", Intell. Fuzzy Syst., 38(3), 2725-2735. https://doi.org/10.3233/JIFS-179558.
  14. Mohammed, A., Burhan, L., Ghafor, K., Sarwar, W. and Mahmood, W. (2021), "Artificial neural network (ANN), M5P-tree, and regression analyses to predict the early age compression strength of concrete modified with DBC-21 and VK-98 polymers", Neural Comput. Appl., 33(13), 7851-7873. https://doi.org/10.1007/s00521-020-05525-y.
  15. Nehdi, M., El Chabib, H. and El Naggar, M.H. (2001), "Predicting performance of self-compacting concrete mixtures using artificial neural networks", Mater. J., 98(5), 394-401.
  16. Neophytou, M.K.A., Pourgouri, S., Kanellopoulos, A.D., Petrou, M.F., Ioannou, I., Georgiou, G. and Alexandrou, A. (2010), "Determination of the rheological parameters of self-compacting concrete matrix using slump flow test", Appl. Rheol., 20(6), 2402-1. https://doi.org/10.3933/applrheol-20-62402.
  17. Ponick, A., Langer, A., Beyer, D., Coenen, M., Haist, M. and Heipke, C. (2022), "Image-based deep learning for rheology determination of bingham fluids", Int. Arch. Photogram. Remote Sens. Spat. Inform. Sci., 43, 711-720. https://doi.org/10.5194/isprs-archives-XLIII-B2-2022-711-2022.
  18. Safayenikoo, H., Khajehzadeh, M. and Nehdi, M.L. (2022), "Novel evolutionary-optimized neural network for predicting fresh concrete slump", Sustainab., 14(9), 4934. https://doi.org/10.3390/su14094934.
  19. Saha, P., Debnath, P. and Thomas, P. (2020), "Prediction of fresh and hardened properties of self-compacting concrete using support vector regression approach", Neural Comput. Appl., 32(12), 7995-8010. https://doi.org/10.1007/s00521-019-04267-w.
  20. Sankar, B. and Ramadoss, P. (2023), "Modeling the compressive strength of high-performance concrete containing metakaolin using distinctive statistical techniques", Result. Control Opt., 12, 100241. https://doi.org/10.1016/j.rico.2023.100241.
  21. Tuan, N.M., Van Hau, Q., Chin, S. and Park, S. (2021), "In-situ concrete slump test incorporating deep learning and stereo vision", Automat. Constr., 121, 103432. https://doi.org/10.1016/j.autcon.2020.103432.
  22. unlu, R. (2020), "An assessment of machine learning models for slump flow and examining redundant features", Comput. Concrete, 25(6), 565-574. https://doi.org/10.12989/cac.2020.25.6.565.
  23. Wijaya, D., Prayogo, D., Santoso, D.I., Gunawan, T. and Widjaja, J.A. (2020), "Optimizing prediction accuracy of concrete mixture behavior using hybrid k-means clustering and ensemble machine learning", J. Phys.: Conf. Ser., 1625(1), 012022. https://doi.org/10.1088/1742-6596/1625/1/012022.
  24. Yang, L., An, X. and Du, S. (2021), "Estimating workability of concrete with different strength grades based on deep learning", Measure., 186, 110073. https://doi.org/10.1016/j.measurement.2021.110073.
  25. Yoon, J., Kim, H., Ju, S., Li, Z. and Pyo, S. (2023), "Framework for rapid characterization of fresh properties of cementitious materials using point cloud and machine learning", Constr. Build. Mater., 400, 132647. https://doi.org/10.1016/j.conbuildmat.2023.132647.
  26. Yoon, J., Kim, H., Sim, S.H. and Pyo, S. (2021), "Framework for characterizing the time-dependent volumetric properties of aerated cementitious material", Constr. Build. Mater., 284, 122781. https://doi.org/10.1016/j.conbuildmat.2021.122781.
  27. Yoon, J., Li, Z. and Kim, H. (2024), "Evaluation of aggregate segregation in self-consolidating concrete using 3D point cloud analysis", Build. Eng., 82, 108199. https://doi.org/10.1016/j.jobe.2023.108199.
  28. Zhang, X., Akber, M.Z. and Zheng, W. (2022), "Predicting the slump of industrially produced concrete using machine learning: A multiclass classification approach", J. Build. Eng., 58, 104997. https://doi.org/10.1016/j.jobe.2022.104997.