한국전산구조공학회논문집 (Journal of the Computational Structural Engineering Institute of Korea)

  • 제37권2호
  • /
  • Pages.143-149
  • /
  • 2024
  • /
  • 1229-3059(pISSN)
  • /
  • 2287-2302(eISSN)
한국전산구조공학회 (Computational Structural Engineering Institute of Korea)
권호 표지
DOI QR코드

DOI QR Code

Unet-VGG16 모델을 활용한 순환골재 마이크로-CT 미세구조의 천연골재 분할

Segmentation of Natural Fine Aggregates in Micro-CT Microstructures of Recycled Aggregates Using Unet-VGG16

  • 홍성욱 (연세대학교 건설환경공학과) ;
  • 문덕기 (연세대학교 건설환경공학과) ;
  • 김세윤 (연세대학교 건설환경공학과) ;
  • 한동석 (연세대학교 건설환경공학과)
  • Sung-Wook Hong (Department of Civil and Environmental Engineering, Yonsei University) ;
  • Deokgi Mun (Department of Civil and Environmental Engineering, Yonsei University) ;
  • Se-Yun Kim (Department of Civil and Environmental Engineering, Yonsei University) ;
  • Tong-Seok Han (Department of Civil and Environmental Engineering, Yonsei University)
  • 투고 : 2024.03.11
  • 심사 : 2024.03.27
  • 발행 : 2024.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

이미지 분석을 통한 재료의 상 구분은 재료의 미세구조 분석을 위해 필수적이다. 이미지 분석에 주로 사용되는 마이크로-CT 이미지는 대체로 재료를 구성하고 있는 상에 따라 회색조 값이 다르게 나타나므로 이미지의 회색조 값 비교를 통해 상을 구분한다. 순환골재의 고체상은 수화된 시멘트풀과 천연골재로 구분되는데, 시멘트풀과 천연골재는 CT이미지 상에서 유사한 회색조 분포를 보여 상을 구분하기 어렵다. 본 연구에서는 Unet-VGG16 네트워크를 활용하여 순환골재 CT 이미지로부터 천연골재를 분할하는 자동화 방법을 제안하였다. 딥러닝 네트워크를 활용하여 2차원 순환골재 CT 이미지로부터 천연골재 영역을 분할하는 방법과 이를 3차원으로 적층하여 3차원 천연골재 이미지를 얻는 방법을 제시하였다. 선별된 3차원 천연골재 이미지에서 각각의 골재 입자를 분할하기 위해 이미지 필터링을 사용하였다. 골재 영역 분할 성능을 정확도, 정밀도, 재현율 F1 스코어를 통해 검증하였다.

Segmentation of material phases through image analysis is essential for analyzing the microstructure of materials. Micro-CT images exhibit variations in grayscale values depending on the phases constituting the material. Phase segmentation is generally achieved by comparing the grayscale values in the images. In the case of waste concrete used as a recycled aggregate, it is challenging to distinguish between hydrated cement paste and natural aggregates, as these components exhibit similar grayscale values in micro-CT images. In this study, we propose a method for automatically separating the aggregates in concrete, in micro-CT images. Utilizing the Unet-VGG16 deep-learning network, we introduce a technique for segmenting the 2D aggregate images and stacking them to obtain 3D aggregate images. Image filtering is employed to separate aggregate particles from the selected 3D aggregate images. The performance of aggregate segmentation is validated through accuracy, precision, recall, and F1-score assessments.

키워드

과제정보

본 연구는 한국연구재단의 지원(NRF_2022R1A4A1033925)의 지원을 받아 수행한 연구 과제입니다.

참고문헌

  1. Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G.S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., Levenberg, J., Mane, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Viegas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., Zheng, X. (2015) Tensorflow: Large-scale Machine Learning on Heterogeneous Distributed Systems, Software available from tensorflow.org.
  2. Bai, C., Shao, L., Da Silva, A.J., Zhao, Z. (2003) A Generalized Model for the Conversion from CT Numbers to Linear Attenuation Coefficients, IEEE Transactions on Nuclear Science, 50(5), pp.1510~1515.
  3. Bai, G., Zhu, C., Liu, C., Liu, B. (2020) An Evaluation of the Recycled Aggregate Characteristics and the Recycled Aggregate Concrete Mechanical Properties, Constr. Build. Mater., 240, p.117978.
  4. Bangaru, S.S., Wang, C., Zhou, X., Hassan, M. (2022) Scanning Electron Microscopy (SEM) Image Segmentation for Microstructure Analysis of Concrete using U-net Convolutional Neural Network. Autom. Constr., 144, p.104602.
  5. Cha, Y.J., Choi, W., Buyukozturk, O. (2017) Deep Learning- based Crack Damage Detection using Convolutional Neural Networks, Comput-Aided Civ Inf, 32(5), pp.361~378.
  6. Chen, Z., Ting, D., Newbury, R., Chen, C. (2021) Semantic Segmentation for Partially Occluded Apple Trees based on Deep Learning, Comput. & Electron. Agric., 181, p.105952.
  7. Chicco, D., Jurman, G. (2020) The Advantages of the Matthews Correlation Coefficient (MCC) over F1 Score and Accuracy in Binary Classification Evaluation, BMC Genom., 21, pp.1~13.
  8. Chung, S.Y., Kim, J.S., Kamm, P.H., Stephan, D., Han, T.S., Abd Elrahman, M. (2021) Pore and Solid Characterizations of Interfacial Transition Zone of Mortar using Microcomputed Tomography Images, J. Mater. Civ. Eng., 33(12), p.04021348.
  9. Goutte, C., Gaussier, E. (2005) A Probabilistic Interpretation of Precision, Recall and F-Score, with Implication for valuation, In: Advances in Information Retrieval: 27th European Conference on IR Research, ECIR 2005, pp.345~359.
  10. Han, T.S., Eum, D., Kim, S.Y., Kim, J.S., Lim, J.H., Park, K., Stephan, D. (2023) Multi-scale Analysis Framework for Predicting Tensile Strength of Cement Paste by Combining Experiments and Simulations, Cem. Concr. Compos., 139, p.105006.
  11. He, K., Zhang, X., Ren, S., Sun, J. (2016) Deep Residual Learning for Image Recognition, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.770~778.
  12. Hu, X., Fang, H., Yang, J., Fan, L., Lin, W., Li, J. (2022) Online Measurement and Segmentation Algorithm of Coarse Aggregate based on Deep Learning and Experimental Comparison, Constr. Build. Mater., 327, 127033.
  13. Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q. (2017) Densely Connected Convolutional Networks, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.4700~4708.
  14. Kim, J.S., Chung, S.Y., Han, T.S., Stephan, D., Abd Elrahman, M. (2020) Correlation between Microstructural Characteristics from Micro-CT of Foamed Concrete and Mechanical Behaviors Evaluated by Experiments and Simulations, Cem. Concr. Compos., 112, p.103657.
  15. Kim, J.S., Kim, J.H., Han, T.S. (2019a) Microstructure Characterization of Cement Paste from Micro-CT and Correlations with Mechanical Properties Evaluated from Virtual and Real Experiments, Mater. Charact., 155, p.109807.
  16. Kim, J.S., Suh, J., Pae, J., Moon, J., Han, T.S. (2022) Gradientbased Phase Segmentation Method for Characterization of Hydrating Cement Paste Microstructures Obtained from X-ray Micro-CT, J. Build. Eng., 46, p.103721.
  17. Kim, S.Y., Kim, J.S., Kang, J.W., Han, T.S. (2019b) Construction of Virtual Interfacial Transition Zone (ITZ) Samples of Hydrated Cement Paste using Extended Stochastic Optimization, Cem. Concr. Compos., 102, pp.84~93. https://doi.org/10.1016/j.cemconcomp.2019.04.012
  18. Liu, Y., Yeoh, J.K. (2021) Robust Pixel-Wise Concrete Crack Segmentation and Properties Retrieval using Image Patches, Autom. Constr, 123, p.103535.
  19. Long, J., Shelhamer, E., Darrell, T. (2015) Fully Convolutional Networks for Semantic Segmentation, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.3431~3440.
  20. Marr, D., Hildreth, E. (1980) Theory of Edge Detection, Proc. Royal Soc. B, 207(1167), pp.187~217.
  21. Meyer, F. (2001) An Overview of Morphological Segmentation, Int. J. Pattern Recognit. Artif. Intell., 15(07), pp.1089~1118.
  22. Ronneberger, O., Fischer, P., Brox, T. (2015) U-net: Convolutional Networks for Biomedical Image Segmentation, In Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015: 18th International Conference, pp.234~241.
  23. Sharma, M.K. (2014) A Survey of Thresholding Techniques over Images, J. Jaipur Nat. Univ., 3(2), pp.461~478. https://doi.org/10.5958/2277-4912.2014.00010.1
  24. Simonyan, K., Zisserman, A. (2014) Very Deep Convolutional Networks for Large-Scale Image Recognition, arXiv preprint arXiv:1409.1556.
  25. The MathWorks Inc. (2023) MATLAB version: 9.13.0 (R2023b), Natick, Massachusetts: The MathWorks Inc. https://www.mathworks.com.
  26. Ullah, M., Mir, J., Husain, S.S., Shahid, M.L.U.R., Ahmad, A. (2024) Concrete Forensic Analysis using Deep Learning-based Coarse Aggregate Segmentation, Autom. Constr., 162, p.105372.
  27. Werner, A.M., Lange, D.A. (1999) Quantitative image analysis of masonry mortar microstructure. J. Comput. Civ. Eng., 13(2), pp.110~115. https://doi.org/10.1061/(ASCE)0887-3801(1999)13:2(110)
  28. Xiao, J., Li, J., Zhang, C. (2005) Mechanical Properties of Recycled Aggregate Concrete under Uniaxial Loading, Cem. Concr. Res., 35(6), pp.1187~1194.
  29. Yang, R., Buenfeld, N.R. (2001) Binary Segmentation of Aggregate in SEM Image Analysis of Concrete, Cem. Concr. Res., 31(3), pp.437~441.
  30. Yasnoff, W.A., Mui, J.K., Bacus, J.W. (1977) Error Measures for Scene Segmentation, Pattern Recognit., 9(4), pp.217~231.