|
|
|
| AI super-resolution reconstruction of nano-CT digital coal-rock |
| ZHANG Liao1, 2, LIU Yintong2, MAO Tingting1, 2, CHENG Jianchao1, 2, HOU Mengdong2, ZHOU Shenghao1, 2, LI Juan2, YAO Jinyue2, XUE Dongjie1, 2, 3? |
(1. State Key Laboratory for Tunnel Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China; 2. School of Mechanics and Civil Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China;
3. State Key Laboratory of Intelligent Coal Mining and Strata Control, China Coal Technology and Engineering Group, Beijing 100013, China) |
|
|
|
|
Abstract Nano-scale coal-rock CT imaging offers the advantages of high resolution and three-dimensional structural reconstruction; however, its acquisition is expensive and its axial resolution is limited, which compromises digital core modeling and seepage simulation. To enhance axial resolution while preserving the true topological structure, this study proposes a generative adversarial network-based super-resolution reconstruction method, termed SRGAN-CT. The proposed method integrates residual blocks and a perceptual loss, and develops a specialized discriminator to strengthen the discrimination of structural details. In addition, a transverse-slice mapping strategy is adopted to construct the training dataset, and orthogonal experiments are conducted to evaluate the robustness and generalization performance of the model. The results show that SRGAN-CT effectively preserves both fine details and topological structures of nano-CT images under multiple upsampling factors and random perturbations, outperforming conventional super-resolution methods in terms of structural-metric deviations and topological errors. Overall, this study verifies the high-fidelity reconstruction capability of SRGAN-CT for 3D topological and geometric information, providing a new technical pathway for efficiently obtaining high-resolution digital cores.
|
|
|
|
|
|
[1] LAWAL A I，KWON S. Application of artificial intelligence to rock mechanics：An overview[J]. Journal of Rock Mechanics and Geotechnical Engineering，2021，13(1)：248–266.
[2] PANAITESCU C T，WU K，TANINO Y，et al. AI enabled digital rock technology for larger scale modelling of complex fractured subsurface rocks[C]// SPE Offshore Europe Conference and Exhibition. SPE，2023：D021S009R001.
[3] SHAIK A R，Al-RATROUT A A，ALSUMAITI A M，et al. Rock classification based on micro-CT images using machine learning techniques[C]// Abu Dhabi International Petroleum Exhibition and Conference. SPE，2019：D021S060R003.
[4] DA W Y，ARMSTRONG R，MOSTAGHIMI P. Super resolution convolutional neural network models for enhancing resolution of rock micro-CT images[J]. arXiv preprint arXiv：1904.07470，2019.
[5] BIZHANI M，ARDAKANI O H，LITTLE E. Reconstructing high fidelity digital rock images using deep convolutional neural networks[J]. Scientific Reports，2022，12(1)：4 264.
[6] OMORI T，SUZUKI S，MICHIBAYASHI K，et al. Super-resolution of X-ray CT images of rock samples by sparse representation：Applications to the complex texture of serpentinite[J]. Scientific Reports，2023，13(1)：6 648.
[7] 陆明远，周宏伟，王 军，等. 深部原位多场环境重构概念下岩石力学行为研究进展[J]. 煤炭学报，2025，50(9)：4 135–4 150.(LU Mingyuan，ZHOU Hongwei，WANG Jun，et al. Research progress on rock mechanical behavior under concept of reconstructed deep in-situ multi-physics conditions[J]. Journal of China Coal Society，2025，50(9)：4 135–4 150.(in Chinese))
[8] 姚明宇，李天娇，丛文雨，等. 页岩数字岩心重构与水力压裂数值试验[J]. 工程科学学报，2024，46(3)：556–566.(YAO Mingyu，LI Tianjiao，CONG Wenyu，et al. Reconstruction of the shale digital core and numerical test of hydraulic fracturing[J]. Chinese Journal of Engineering，2024，46(3)：556–566.(in Chinese))
[9] 张 良，齐庆新，REN Ting，等. 基于显微CT扫描和统计强度的煤岩损伤破裂特性研究[J]. 煤炭科学技术，2023，51(增2)：1–12.(ZHANG Liang，QI Qingxin，REN Ting，et al. Study on the damage and fracture characteristics of coal rock based on the X-ray micro-CT scanning technology and statistical strength theory[J]. Coal Science and Technology，2023，51(Supp.2)：1–12.(in Chinese))
[10] 朱凯然，王 波，武晓朦，等. 原位岩土CT及其关键技术探讨[J]. CT理论与应用研究，2019，28(6)：641–652.(ZHU Kairan, WANG Bo, WU Xiaomeng，et al. In-situ geotechnical CT and its key technologies[J]. Computerized Tomography Theory and Application，2019，28(6)：641–652.(in Chinese))
[11] YOU S，LIAO Q，YAN Z，et al. Super-resolution reconstruction of 3D digital rocks by deep neural networks[J]. Geoenergy Science and Engineering，2024，237：212781.
[12] WANG Y D，ARMSTRONG R T，MOSTAGHIMI P. Boosting resolution and recovering texture of 2D and 3D micro-CT images with deep learning[J]. Water Resources Research，2020，56(1)：e2019WR026052.
[13] JIANG F，GUO Y，TSUJI T，et al. Upscaling permeability using multiscale X-ray-CT images with digital rock modeling and deep learning techniques[J]. Water Resources Research，2023，59(3)：e2022WR033267.
[14] YANG Y，WANG J，LIU F，et al. Super-resolution of digital rock images with hybrid attention multi-branch neural network[J]. Gas Science and Engineering，2024，128：205395.
[15] SHAN L，LIU C，LIU Y，et al. Rock CT image super-resolution using residual dual-channel attention generative adversarial network[J]. Energies，2022，15(14)：5 115.
[16] LIU M，MUKERJI T. Multiscale fusion of digital rock images based on deep generative adversarial networks[J]. Geophysical Research Letters，2022，49(9)：e2022GL098342.
[17] CASTRO V F J，DUARTE V A，CARVALHO M L，et al. Absolute permeability estimation from microtomography rock images through deep learning super-resolution and adversarial fine tuning[J]. Scientific Reports，2024，14(1)：16 704.
[18] UGOLKOV E，HE X，KWAK H，et al. Super-resolution of 3D micro-CT images using generative adversarial networks：enhancing resolution and segmentation accuracy[J]. arXiv preprint arXiv：2501.06939，2025.
[19] SHAN L，BAI X，LIU C，et al. Super-resolution reconstruction of digital rock CT images based on residual attention mechanism[J]. Advances in Geo-Energy Research，2022，6(2)：157–168.
[20] ROSLIN A，MARSH M，PICHÉ N，et al. Processing of micro-CT images of granodiorite rock samples using convolutional neural networks (CNN)，Part I：Super-resolution enhancement using a 3D CNN[J]. Minerals Engineering，2022，188：107748.
[21] MA Z，SUN S，YAN B，et al. Enhancing the resolution of micro-CT images of rock samples via unsupervised machine learning based on a diffusion model[C]// SPE Annual Technical Conference and Exhibition. SPE，2023：D021S028R005.
[22] LIU C，LIU Y，SHAN L，et al. Enhancing unsupervised rock CT image super-resolution with non-local attention[J]. Geoenergy Science and Engineering，2024，238：212912.
[23] ALQAHTANI N J，NIU Y，WANG Y D，et al. Super-resolved segmentation of X-ray images of carbonate rocks using deep learning[J]. Transport in Porous Media，2022，143(2)：497–525.
[24] ZHAO B，SAXENA N，HOFMANN R，et al. Enhancing resolution of micro-CT images of reservoir rocks using super resolution[J]. Computers and Geosciences，2023，170：105265.
[25] 贾 胜，张 村，吴润泽，等. 基于纳米CT扫描重构的含水层砂岩流–固耦合分析[J]. 岩石力学与工程学报，2026，45(1)：144–154.(JIA Sheng，ZHANG Cun，WU Runze，et al. Fluid-solid coupling analysis of aquifer sandstone based on Nano-CT scanning reconstruction[J]. Chinese Journal of Rock Mechanics and Engineering，2026，45(1)：144–154.(in Chinese))
[26] HINTON G E，SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science，2006，313(5786)：504–507.
[27] 薛东杰，唐麒淳，王 傲，等. 煤岩微观相态FCN智能识别与分形重构[J]. 岩石力学与工程学报，2020，39(6)：1 203–1 221.(XUE Dongjie，TANG Qichun，WANG Ao，et al. FCN-based intelligent identification and fractal reconstruction of pore-fracture network in coal by micro CT scanning[J]. Chinese Journal of Rock Mechanics and Engineering，2020，39(6)：1 203–1 221.(in Chinese))
[28] 薛东杰，唐麒淳，王 傲，等. 基于FCN的岩石混凝土裂隙几何智能识别[J]. 岩石力学与工程学报，2019，38(增2)：3 393–3 403.(XUE Dongjie，TANG Qichun，WANG Ao，et al. FCN-based intelligent identification of crack geometry in rock or concrete[J]. Chinese Journal of Rock Mechanics and Engineering，2019，38(Supp.2)：3 393–3 403.(in Chinese))
[29] 李 刚，张亚兵，杨庆贺，等. 基于Real-ESRGAN的岩石CT图像超分辨率重建[J]. 工矿自动化，2023，49(11)：84–91.(LI Gang， ZHANG Yabing，YANG Qinghe，et al. Super-resolution reconstruction of rock CT images based on Real-ESRGAN[J]. Journal of Mine Automation，2023，49(11)：84–91.(in Chinese))
[30] 邹 亮，陈志锋，谭智毅，等. 基于深度神经网络的煤岩显微图像划痕检测与修复[J]. 煤炭科学技术，2023，51(增2)：275–284.(ZOU Liang，CHEN Zhifeng，TAN Zhiyi，et al. Scratch detection and restoration of coal photomicrograph via deep neural network[J]. Coal Science and Technology，2023，51(Supp.2)：275–284.(in Chinese))
[31] GOODFELLOW I J，POUGET-ABADIE J，MIRZA M，et al. Generative adversarial nets[J]. Advances in Neural Information Processing Systems，2014，27：2 672–2 680.
[32] RADFORD A，METZ L，CHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. arXiv preprint arXiv：1511.06434，2015.
[33] KARRAS T，AILA T，LAINE S，et al. Progressive growing of gans for improved quality，stability，and variation[J]. arXiv preprint arXiv：1710.10196，2017.
[34] BROCK A，DONAHUE J，SIMONYAN K. Large scale GAN training for high fidelity natural image synthesis[J]. arXiv preprint arXiv：1809.11096，2018.
[35] GULRAJANI I，AHMED F，ARJOVSKY M，et al. Improved training of Wasserstein GANs[J]. Advances in neural information processing systems，2017，30：5 767–5 777.
[36] SAXENA D，CAO J. Generative adversarial networks (GANs) challenges，solutions，and future directions[J]. ACM Computing Surveys (CSUR)，2021，54(3)：1–42.
[37] LEDIG C，THEIS L，HUSZÁR F，et al. Photo-realistic single image super-resolution using a generative adversarial network[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]：[s. n.]，2017：4 681–4 690.
[38] ISOLA P，ZHU J Y，ZHOU T，et al. Image-to-image translation with conditional adversarial networks[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]：[s. n.]，2017：1 125–1 134.
[39] WANG X，YU K，WU S，et al. ESRGAN: Enhanced super-resolution generative adversarial networks[C]//Proceedings of the 15th European Conference on Computer Vision (ECCV) Workshops. Munich：Springer，2018：63–79.
[40] JOHNSON J，ALAHI A，FEI-FEI L. Perceptual losses for real-time style transfer and super-resolution[C]// European Conference on Computer Vision. Cham：Springer International Publishing，2016：694–711.
[41] ZHENG T，ODA H，MORIYA T，et al. Multi-modality super-resolution loss for GAN-based super-resolution of clinical CT images using micro CT image database[C]// Medical Imaging 2020：Image Processing. SPIE，2020，11313：7–13.
[42] SHEN Y，LIANG N，ZHONG X，et al. CT image super-resolution under the guidance of deep gradient information[J]. Journal of X-Ray Science and Technology，2025，33(1)：58–71.
[43] CHEN H，HE X，TENG Q，et al. Super-resolution of real-world rock microcomputed tomography images using cycle-consistent generative adversarial networks[J]. Physical Review E，2020，101(2)：023305.
[44] 尹梦晓，林振峰，杨 锋. 基于动态感受野的自适应多尺度信息融合的图像转换[J]. 电子与信息学报，2021，43(8)：2 386–2 394.(YIN Mengxiao，LIN Zhenfeng，YANG Feng. Adaptive multi-scale information fusion based on dynamic receptive field for image-to-image translation[J]. Journal of Electronics and Information Technology，2021，43(8)：2 386–2 394.(in Chinese))
[45] UHM K H，CHO H，HONG S H，et al. An anisotropic cross-view texture transfer with multi-reference non-local attention for CT slice interpolation[J]. IEEE Transactions on Medical Imaging，2026，45(1)： 336–349.
[46] CHEN G，ZHANG G，YANG Z，et al. Multi-scale patch-GAN with edge detection for image inpainting[J]. Applied Intelligence，2023，53(4)：3 917–3 932.
[47] 何鑫睿，李秀梅，孙军梅，等. 基于改进Pix2Vox的单图像三维重建网络[J]. 计算机辅助设计与图形学学报，2022，34(3)：364–372.(HE Xinrui，LI Xiumei，SUN Junmei，et al. Improved Pix2Vox based 3D reconstruction network from single image[J]. Journal of Computer-Aided Design and Computer Graphics，2022，34(3)：364–372.(in Chinese))
[48] CHOI H，LEE J. Efficient use of GPU memory for large-scale deep learning model training[J]. Applied Sciences，2021，11(21)：10377.
[49] PASZKE A，GROSS S，MASSA F，et al. PyTorch：An imperative style，high-performance deep learning library[J]. Advances in Neural Information Processing Systems，2019，32.
[50] GAVALI P，BANU J S. Deep convolutional neural network for image classification on CUDA platform[M]. [S. l.]：Academic Press，2019：99–122.
[51] GHOSH A，NAYAK A，PANWAR A，et al. PyGraph：Robust compiler support for CUDA graphs in PyTorch[J]. arXiv preprint arXiv：2503.19779，2025.
[52] ARABBOEV M，BEGMATOV S，RIKHSIVOEV M，et al. A comprehensive review of image super-resolution metrics：classical and AI-based approaches[J]. Acta IMEKO，2024，13(1)：1–8.
[53] SHAN L，LIU C，LIU Y，et al. Single image multi-scale enhancement for rock micro-CT super-resolution using residual U-Net[J]. Applied Computing and Geosciences，2024，22：100165.
[54] KIM H，LEE H，LEE D. Deep learning-based computed tomographic image super-resolution via wavelet embedding[J]. Radiation Physics and Chemistry，2023，205：110718.
[55] KIM D，HAN D，KIM M，et al. Evaluation of AI-based super-resolution model in CT for dose reduction method：a phantom study[J]. Egyptian Journal of Radiology and Nuclear Medicine，2025，56(1)：159.
[56] ZHANG R，ISOLA P，EFROS A A，et al. The unreasonable effectiveness of deep features as a perceptual metric[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]：[s. n.]，2018：586–595.
[57] GHILDYAL A，LIU F. Shift-tolerant perceptual similarity metric[C]// European Conference on Computer Vision. Cham：Springer Nature Switzerland，2022：91–107.
[58] DONG C，LOY C C，HE K，et al. Image super-resolution using deep convolutional networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，38(2)：295–307.
[59] LIM B，SON S，KIM H，et al. Enhanced deep residual networks for single image super-resolution[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. [S. l.]：[s. n.]，2017：136–144. |
|
|
|
2026, Vol. 45 
