Abstract
Geotechnical particle image velocimetry (GeoPIV), as a type of digital image correlation (DIC), represents the state-of-the-art methodology for non-contact full-field deformation measurement in geotechnical engineering. Yet, when applying GeoPIV on sand specimens with interests in grain level, the discontinuities detection at grain boundaries remains as a challenge for 2D GeoPIV applications. In order to facilitate the full-field measurement for microscopic study, a method is proposed in this study to realize 2D pixel-level motion calculation using supervised optical flow algorithm with deep networks. Using digital images acquired from direct shear testing, the performance of this approach is demonstrated and compared with the prevailing GeoPIV method. Two series of experiments using small and large displacement modes were conducted, respectively, to demonstrate the method’s ability of revealing greater insights on soil behavior at grain level. To verify its accuracy, performance benchmarking of the approach was also conducted. Besides, a method was proposed to evaluate the errors in experimental images to ensure the accuracy and precision. It was demonstrated that the proposed method can achieve accurate pixel-level motion field calculation using images of common size and that the deformation discontinuities among particles can be clearly presented.
Similar content being viewed by others
Data availability
Some or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.
References
Alshibli KA, Sture S (2000) Shear band formation in plane strain experiments of sand. J Geotech Geoenviron Eng 126(6):495–503
Bai M, Luo W, Kundu K, Urtasun R (2016) Exploiting semantic information and deep matching for optical flow. European Conference on Computer Vision, Springer: 154–170.
Boukhtache S, Abdelouahab K, Berry F, Blaysat B, Grediac M, Sur F (2021) When deep learning meets digital image correlation. Opt Lasers Eng 136:106308
Chen Z, Lenthe W, Stinville J, Echlin M, Pollock T, Daly S (2018) High-resolution deformation mapping across large fields of view using scanning Electron microscopy and digital image correlation. Exp Mech 58(9):1407–1421
Cheng X, Zhou S, Xing T, Zhu Y, Ma S (2023) Solving digital image correlation with neural networks constrained by strain-displacement relations. Opt Express 31(3):3865–3880
Chivers K, Clocksin W (2000) Inspection of surface strain in materials using optical flow. BMVC, Citeseer, pp 1–10
DeJong JT, Westgate ZJ (2009) Role of initial state, material properties, and confinement condition on local and global soil-structure interface behavior. J Geotech Geoenviron Eng 135(11):1646–1660
Dosovitskiy A, Fischer P et al. (2015) Flownet: learning optical flow with convolutional networks. In: Proceedings of the IEEE international conference on computer vision, pp 2758–2766
Feia S, Sulem J, Canou J, Ghabezloo S, Clain X (2016) Changes in permeability of sand during triaxial loading: effect of fine particles production. Acta Geotech 11:1–19
Fortun D, Bouthemy P, Kervrann C (2015) Optical flow modeling and computation: a survey. Comput Vis Image Underst 134:1–21
Gong X, Bansmer S (2015) Horn–Schunck optical flow applied to deformation measurement of a birdlike airfoil. Chin J Aeronaut 28(5):1305–1315
Hall SA (2012) Digital image correlation in experimental geomechanics. ALERT Geomater Doctor Summer School 2012:69–102
Hartmann C, Volk W (2019) Digital image correlation and optical flow analysis based on the material texture with application on high-speed deformation measurement in shear cutting. In: International conference on digital image and signal processing (DISP 2019), At Oxford, UK.
Horn BK, Schunck BG (1981) Determining optical flow. Artif Intell 17(1–3):185–203
Huang F, Wu C, Ni P, Wan G, Zheng A, Jang B-A, Karekal S (2020) Experimental analysis of progressive failure behavior of rock tunnel with a fault zone using non-contact DIC technique. Int J Rock Mech Min Sci 132:104355
Hur J, Roth S (2020) Optical flow estimation in the deep learning age. In: Modelling human motion: from human perception to robot design, pp 119–140
Ilg E, Mayer N, Saikia T, Keuper M, Dosovitskiy A, Brox T (2017) Flownet 2.0: evolution of optical flow estimation with deep networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2462–2470
Ilg ET, Saikia MK, Brox T (2018) Occlusions, motion and depth boundaries with a generic network for disparity, optical flow or scene flow estimation. In: Proceedings of the European Conference on Computer Vision (ECCV), pp 614–630
Iskander M (2010) Modelling with transparent soils: visualizing soil structure interaction and multi phase flow, non-intrusively. Springer Science & Business Media
Iskander MG, Liu J (2005) Discussions and closures. J Comput Civil Eng 217
Kavazanjian E, Andresen J, Gutierrez A (2017) Experimental evaluation of HDPE geomembrane seam strain concentrations. Geosynth Int 24(4):333–342
Krizhevsky A, Sutskever I, Hinton GE (2017) Imagenet classification with deep convolutional neural networks. Commun ACM 60(6):84–90
Lambe TW, Whitman RV (1991) Soil mechanics. John Wiley & Sons
Lashkari A, Jamali V (2021) Global and local sand–geosynthetic interface behaviour. Géotechnique 71(4):346–367
Lucas BD, Kanade T (1981) An iterative image registration technique with an application to stereo vision. Vancouver
Marshall AM, Klar A, Mair R (2010) Tunneling beneath buried pipes: view of soil strain and its effect on pipeline behavior. J Geotech Geoenviron Eng 136(12):1664–1672
Min H-G, On H-I, Kang D-J, Park J-H (2019) Strain measurement during tensile testing using deep learning-based digital image correlation. Meas Sci Technol 31(1):015014
Nesi P (1993) Variational approach to optical flow estimation managing discontinuities. Image Vis Comput 11(7):419–439
Omidvar M, Chen Z, Iskander M (2015) Image-based Lagrangian analysis of granular kinematics. J Comput Civ Eng 29(6):04014101
Ranjan A, Black MJ (2017) Optical flow estimation using a spatial pyramid network. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4161–4170
Raychowdhury P (2008) Nonlinear Winkler-based shallow foundation model for performance assessment of seismically loaded structures. University of California, San Diego
Rechenmacher AL, Abedi S, Chupin O, Orlando AD (2011) Characterization of mesoscale instabilities in localized granular shear using digital image correlation. Acta Geotech 6:205–217
Rechenmacher AL, Finno RJ (2004) Digital image correlation to evaluate shear banding in dilative sands. Geotech Test J 27(1):13–22
Roscoe KH (1970) The influence of strains in soil mechanics. Geotechnique 20(2):129–170
Sadek S, Iskander MG, Liu J (2003) Accuracy of digital image correlation for measuring deformations in transparent media. J Comput Civ Eng 17(2):88–96
Sadrekarimi A, Olson SM (2010) Shear band formation observed in ring shear tests on sandy soils. J Geotech Geoenviron Eng 136(2):366–375
Stanier SA, Blaber J, Take WA, White D (2016) Improved image-based deformation measurement for geotechnical applications. Can Geotech J 53(5):727–739
Sun D, Yang X, Liu M-Y, Kautz J (2018) Pwc-net: Cnns for optical flow using pyramid, warping, and cost volume. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 8934–8943
Sun D, Yang X, Liu M-Y, Kautz J (2019) Models matter, so does training: an empirical study of cnns for optical flow estimation. IEEE Trans Pattern Anal Mach Intell 42(6):1408–1423
Take WA (2015) Thirty-sixth canadian geotechnical colloquium: advances in visualization of geotechnical processes through digital image correlation. Can Geotech J 52(9):1199–1220
Teed Z, Deng J (2020) Raft: recurrent all-pairs field transforms for optical flow. In: European conference on computer vision, Springer, pp 402–419.
Tovar-Valencia RD, Galvis-Castro A, Salgado R, Prezzi M (2018) Effect of surface roughness on the shaft resistance of displacement model piles in sand. J Geotech Geoenviron Eng 144(3):04017120
Tu Z, Xie W, Zhang D, Poppe R, Veltkamp RC, Li B, Yuan J (2019) A survey of variational and CNN-based optical flow techniques. Signal Process Image Commun 72:9–24
Viggiani G, Hall SA (2012) Full-field measurements in experimental geomechanics: historical perspective, current trends and recent results. ALERT Doctoral School: 3–68.
Wang P, Sang Y, Shao L, Guo X (2019) Measurement of the deformation of sand in a plane strain compression experiment using incremental digital image correlation. Acta Geotech 14:547–557
White D, Take W (2002) GeoPIV: particle image velocimetry (PIV), CUED/D-SOILS/TR322, 1–14
White D, Take W, Bolton M (2003) Soil deformation measurement using particle image velocimetry (PIV) and photogrammetry. Geotechnique 53(7):619–631
Yang R, Li Y, Zeng D, Guo P (2022) Deep DIC: Deep learning-based digital image correlation for end-to-end displacement and strain measurement. J Mater Process Technol 302:117474
Zhang Z, Yin X, Yan Z (2022) Rapid data annotation for sand-like granular instance segmentation using mask-RCNN. Autom Constr 133:103994
Zhang L, Zhang L, Tang W (2008) Similarity of soil variability in centrifuge models. Can Geotech J 45(8):1118–1129
Zhao C, Fauzi UJ (2022) Visualized liquefaction behavior of sandy soil deposited in water under undrained cyclic shearing. Acta Geotech 17(8):3143–3160
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: application of the proposed method on a penetration test with fine sand
Appendix: application of the proposed method on a penetration test with fine sand
1.1 Experimental setting
To evaluate the performance of the proposed method on different types of sand and testing scenario, we conducted another penetration test using finer sand. Figure
16a shows the experimental setup, which was similar to the one used in the main text. A metal plate penetrated into the sand sample, and the front view of the visual window is displayed in Fig. 16c. Figure 16b and Table
4 compare the properties of the finer sand and the sand used in the main text.
1.2 Results
The calculation results are shown in a manner similar to the main text. The penetrating interval is 0.1 mm. Figure
17 shows the incremental displacement field under 0.1 mm shearing in Fig. 17a–c and the accumulated displacement field under 0.2 mm and 0.3 mm shearing in Fig. 17d and e.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhang, Z., Rahardjo, H., Yan, Z. et al. 2D full-field deformation measurement at grain level using optical flow with deep networks. Acta Geotech. (2024). https://doi.org/10.1007/s11440-024-02242-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11440-024-02242-x