Abstract
Rock quality evaluation is an important step for preliminary design and survey in hydraulic engineering. In current practice, borehole televiewer (BHTV) images are manually reviewed and analyzed by domain experts to evaluate rock quality, which is time-consuming, laborious, and prone to the subjectivity of human evaluation. Emerging techniques, such as deep learning and image processing, can potentially address the limitations by automating the process of BHTV image analysis. In this research, we propose an intelligent image segmentation model based on ResNet and Unet, which is called RUnet. In the model, ResNet is used as the pre-trained model to extract image features. The optimizer and the loss function of the RUNet are improved by incorporating the domain knowledge of geology. Through the comparison of the Unet and RUnet with different optimizers, the effectiveness of the model has been validated. Based on the features extracted by RUnet, the binary image skeleton can be obtained and the relative coordinates of all the pixels can be calculated, which can be applied in dip azimuth, dip angle, and mean thickness calculation of the fracture. The generalized least squares method is also employed in fracture occurrence quantification. The intelligent fracture information quantification can be realized through the whole analysis process, which provides an automated and reliable approach to quantitatively evaluate rock quality via BHTV images.
Similar content being viewed by others
Code availability
Source codes can be found at GitHub https://github.com/ye3010205121/Quantitative-Analysis-of-Geological-Fracture.
References
Al-Sit W, Al-Nuaimy W, Marelli M, Al-Ataby A (2015) Visual texture for automated characterisation of geological features in borehole televiewer imagery. J Appel Geophys 119:139–146
Avendi MR, Kheradvar A, Jafarkhani H (2016) A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI. Med Image Anal 30:108–119
Bae D, Kim K, Koh Y, Kim J (2011) Characterization of joint roughness in granite by applying the scan circle technique to images from a borehole televiewer. Rock Mech Rock Eng 44:497–504
Barton N, Lien R, Lunde J (1974) Engineering classification of rock masses for the design of tunnel support. Rock Mech Rock Eng 6(4):189–239
Bieniawski ZT (1989) Engineering rock mass classification. Wiley, New York
Bieniek A, Moga A (2000) An efficient watershed algorithm based on connected components. Pattern Recogn 33(6):907–916
Bottou L (2010) Large-scale machine learning with stochastic gradient descent. In: proceedings of COMPSTAT'2010. Heidelberg: Physica-Verlag HD 177–186
Cha YJ, Choi W, Suh G, Mahmoudkhani S, Büyüköztürk O (2018) Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types. Comput-Aided Civ Inf 33(9):731–747
Chai H, Li N, Xiao C, Liu X, Wang C, Wu D (2009) Automatic discrimination of sedimentary facies and lithologies in reef-bank reservoirs using borehole image logs. Appl Geophys 6:17–29
Chen J, Liu D (2021) Bottom-up image detection of water channel slope damages based on superpixel segmentation and support vector machine. Adv Eng Inform 47: 101205
Chen J, Lu W, Yuan L, Wu Y (2022) Estimating construction waste truck payload volume using monocular vision. Resour Conserv Recy 177: 106013
Chen Z, Liu X, Yang J, Little E, Zhou Y (2020) Deep learning-based method for SEM image segmentation in mineral characterization, an example from Duvernay Shale samples in Western Canada Sedimentary Basin. Comput Geosci 138: 104450
Coudray N, Ocampo PS, Sakellaropoulos T et al (2018) Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning. Nat Med 24:1559–1567
Deere D (1988) The rock quality designation (RQD) index in practice. In: L. Kirkaldie (ed) Rock classification systems for engineering purposes. ASTM International, West Conshohocken, pp 90–101
Deng J, Dong W, Socher R, Li LJ, Li K, Fei-Fei L (2009) Imagenet: a large-scale hierarchical image database. IEEE Computer Society, PROC CVPR IEEE. Piscataway, pp 248–255
Dias LO, Bom CR, Faria EL et al (2020) Automatic detection of fractures and breakouts patterns in acoustic borehole image logs using fast-region convolutional neural networks. J Petrol Sci Eng 191: 107099
Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, Thrun S (2017) Dermatologist-level classification of skin cancer with deep neural networks. Nature 542:115–118
Guo Y, Liu Y, Oerlemans A, Lao S, Wu S, Lew MS (2016) Deep learning for visual understanding: a review[J]. Neurocomputing 187:27–48
Han S, Li H, Li M, Luo X (2019a) Measuring rock surface strength based on spectrograms with deep convolutional networks. Comput Geosci 133: 104312
Han S, Li H, Li M, Rose T (2019) A deep learning based method for the non-destructive measuring of rock strength through hammering sound. Appl Sci-Basel 9(17):3484
He KM, Zhang XY, Ren SQ, Sun J (2016) Deep residual learning for image recognition. PROC CVPR IEEE. Piscataway: IEEE Computer Society 770–778
Hinton G, Srivastava N, Swersky K (2012) Lecture 6d- a separate, adaptive learning rate for each connection. Slides of lecture neural networks for machine learning
Hoek E, Brown ET (1997) Practical estimates or rock mass strength. Int J Rock Mech Min Sci Geomech Abstr 34(8):1165–1186
Huang H, Li Q, Zhang D (2018) Deep learning based image recognition for crack and leakage defects of metro shield tunnel. Tunn Undergr Sp Tech 77:166–176
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980
Kooi T, Litjens G, Van Ginneken B et al (2017) Large scale deep learning for computer aided detection of mammographic lesions. Med Image Anal 35:303–312
Lai J, Wang G, Fan Z, Wang Z, Chen J, Zhou Z, Wang S, Xiao C (2017) Fracture detection in oil-based drilling mud using a combination of borehole image and sonic logs. Mar Petrlo Geol 84:195–214
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521:436–444
Li L, Yu C, Han Z, Sun T (2019) Automatic identification of the rock-soil interface and solution fissures from optical borehole images based on color features. IEEE J-STARS 12:3862–3873
Li D, Yuan R, Ding Z, Xu R (2021) Automatic calculating grain size of gravels based on micro-resistivity image of well. Arab J Geosc 14(17):1–10
Liang X (2019) Image-based post-disaster inspection of reinforced concrete bridge systems using deep learning with Bayesian optimization. Comput-Aided Civ Inf 34:415–430
Liu D, Chen J, Hu D, Zhang Z (2019a) Dynamic BIM-augmented UAV safety inspection for water diversion project. Comput Ind 108:163–177
Liu C, Li M, Zhang Y, Han S, Zhu Y (2019b) An enhanced rock mineral recognition method integrating a deep learning model and clustering algorithm. Minerals 9(9):516
Litjens G, Kooi T, Bejnordi BE, Setio AAA (2017) A survey on deep learning in medical image analysis. Med Image Anal 42:60–88
Long J, Shelhamer E, Darrell T (2015) Fully convolutional networks for semantic segmentation. In: IEEE International conference on computer vision and pattern recognition. PROC CVPR IEEE. Piscataway: IEEE computer society 3431–3440
Ma W, Xu F (2021) Underwater image segmentation based on computer vision and research on recognition algorithm. Arab J Geosci 14(18):1–11
Marangio P, Christodoulou V, Filgueira R, Rogers HF, Beggan CD (2020) Automatic detection of Ionospheric Alfvén Resonances in magnetic spectrograms using U-net. Comput Geosci 145: 104598
Otsu N (1979) A threshold selection method from gray-level histograms. IEEE T Syst Man CY-S 9(1):62–66
Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. Proceedings of international conference on medical image computing and computer-assisted intervention. Cham: Springer 234–241
Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural Netw 61:85–117
Voulodimos A, Doulamis N, Doulamis A, Protopapadakis E (2018) Deep learning for computer vision: a brief review. Comput Intel Neurosc: 2018
Wang C, Zou X, Han Z, Wang Y, Wang J (2016) An automatic recognition and parameter extraction method for structural planes in borehole image. J Appl Geophys 135:135–143
Wang C, Zou X, Han Z, Wang J, Wang Y (2017) The automatic interpretation of structural plane parameters in borehole camera images from drilling engineering. J Petrol Sci Eng 154:417–424
Williams JH, Johnson CD (2004) Acoustic and optical borehole-wall imaging for fractured-rock aquifer studies. J Appl Geophys 55:151–159
Xiong Y, Zuo R (2016) Recognition of geochemical anomalies using a deep autoencoder network. Comput Geosci 86:75–82
Xue Y, Cai X, Shadabfar M, Shao H, ZHAG S (2020) Deep learning-based automatic recognition of water leakage area in shield tunnel lining. Tunn Undergr Sp Tech 104: 103524
Zhang R, Shen J, Wei F, Li X, Sangaiah AK (2017) Medical image classification based on multi-scale non-negative sparse coding. Artif Intell Med 83:44–51
Zhang W, Feng XT, Bi X, Yao ZH, Xiao YX, Hu L, Niu WJ, Feng GL (2021) An arrival time picker for microseismic rock fracturing waveforms and its quality control for automatic localization in tunnels. Comput Geotech 135: 104175
Zhao S, Wu N, Wang Q (2020) Deep residual U-net with input of static structural responses for efficient U* load transfer path analysis. Adv Eng Inf 46: 101184
Zohreh M, Junin R, Jeffreys P (2014) Evaluate the borehole condition to reduce drilling risk and avoid potential well bore damages by using image logs. J Petrol Sci Eng 122:318–330
Zou X, Song H (2021) The fast formation of high-precision panoramic image for the processing of borehole camera video of deep rock mass structures. B Eng Geol Environ 80(3):2199–2213
Acknowledgements
This research was supported by the National Natural Science Foundation of China (NSFC) (Grant No. 52009109, 51979224) and the PhD Research Startup Foundation of Xi’an University of Technology (Grant No. 104-451120005).
Author information
Authors and Affiliations
Contributions
Ye Zhang and Yanlong Li: idea, framework design, coding, and writing; Jinqiao Chen: data processing and calculating.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Communicated by Zeynal Abiddin Erguler.
Responsible Editor: Zeynal Abiddin Erguler
Rights and permissions
About this article
Cite this article
Zhang, Y., Chen, J. & Li, Y. Segmentation and quantitative analysis of geological fracture: a deep transfer learning approach based on borehole televiewer image. Arab J Geosci 15, 300 (2022). https://doi.org/10.1007/s12517-022-09536-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12517-022-09536-y