Application of a generative adversarial network to super-resolution reconstruction of fracture images in logging-while-drilling electrical resistivity imaging

Authors

HAN Sen, No.9 Oil Production Plant, Changqing Oilfield Company, PetroChina, Yinchuan 750005, ChinaFollow
SHEN Chuanhai, No.9 Oil Production Plant, Changqing Oilfield Company, PetroChina, Yinchuan 750005, China
XU Chen, No.9 Oil Production Plant, Changqing Oilfield Company, PetroChina, Yinchuan 750005, China
TIAN Yimeng, No.9 Oil Production Plant, Changqing Oilfield Company, PetroChina, Yinchuan 750005, China
ZHAO Biao, No.9 Oil Production Plant, Changqing Oilfield Company, PetroChina, Yinchuan 750005, China
WANG Ze, No.9 Oil Production Plant, Changqing Oilfield Company, PetroChina, Yinchuan 750005, China
LIU Yongkang, College of Advanced Manufacturing, Nanchang University, Nanchang 330031, China
KANG Zhengming, School of Electronic Engineering, Xi’an Shiyou University, Xi’an 710065, ChinaFollow

Abstract

Objective Logging-while-drilling electrical resistivity imaging (LWD-ERI) enables the intuitive presentation of geobodies near wellbores in images, representing an important approach to the fine-scale evaluation of stratigraphic parameters while also providing significant geological guidance during the drilling of fractured reservoirs. However, this technique faces the challenge of low image resolution in real-time data processing due to its limited well-to-surface data transmission rates, exerting adverse impacts on the qualitative identification of wellbore fractures and the quantitative evaluation of their parameters. To address this issue, this study proposed a super-resolution image reconstruction method based on a generative adversarial network (GAN), aiming to achieve high-definition fracture images. Methods Based on the basic GAN architecture, a 23-layer deep generator network was constructed by integrating the GAN with residual dense blocks (RDBs) and channel attention mechanism (CAM). Meanwhile, a dataset consisting of high-resolution images stored and low-resolution real-time images was constructed using measured data from a resistivity-at-bit (RAB) tool. Subsequently, through training of the GAN using the constructed image dataset, the batch size and learning rate were optimized. Accordingly, the network parameters corresponding to smaller errors and higher precision were determined. Finally, the trained network model was utilized for the super-resolution image reconstruction of real-time data to achieve an image resolution close to that of the stored data. Results and Conclusions Compared to traditional methods, the proposed super-resolution image reconstruction method improved the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) index by 2.2 dB and 0.6, respectively. Under 4× downsampling, the intelligent method remained effective in reconstructing the image features of fractured geobodies, significantly enhancing the resolution of real-time LWD images. The results of this study hold great significance for enhancing the real-time geological guidance during drilling operations.

Keywords

logging while drilling (LWD), electrical resistivity imaging (ERI), fracture, generative adversarial network (GAN), super-resolution reconstruction

DOI

10.12363/issn.1001-1986.25.08.0645

Recommended Citation

HAN Sen, SHEN Chuanhai, XU Chen, et al. (2026) "Application of a generative adversarial network to super-resolution reconstruction of fracture images in logging-while-drilling electrical resistivity imaging," Coal Geology & Exploration: Vol. 54: Iss. 3, Article 23.
DOI: 10.12363/issn.1001-1986.25.08.0645
Available at: https://cge.researchcommons.org/journal/vol54/iss3/23

Reference

[1] YING Dawang,BLUNT M J,ARMSTRONG R T,et al. Deep learning in pore scale imaging and modeling[J]. Earth–Science Reviews,2021,215:103555.

[2] SIU W C，HUNG K W. Review of image interpolation and super–resolution[C]//Proceedings of the 2012 Asia Pacific Signal and Information Processing Association Annual Summit and Conference. Hollywood：IEEE，2012：1–10.

[3] NASONOV A V，KRYLOV A S. Fast super–resolution using weighted median filtering[C]//2010 20th International Conference on Pattern Recognition. Istanbul：IEEE，2010：2230–2233.

[4] WANG Jingwei，ZHOU Peng，HAN Xianjun，et al. Medical image super–resolution via diagnosis–guided attention[C]//2023 IEEE International Conference on Multimedia and Expo (ICME). Brisbane：IEEE，2023：462–467.

[5] CHEN Keyan,LI Wenyuan,LEI Sen,et al. Continuous remote sensing image super–resolution based on context interaction in implicit function space[J]. IEEE Transactions on Geoscience and Remote Sensing,2023,61:4702216.

[6] DONG Xiaoyu,SUN Xu,JIA Xiuping,et al. Remote sensing image super–resolution using novel dense–sampling networks[J]. IEEE Transactions on Geoscience and Remote Sensing,2021,59(2):1618−1633.

[7] DONG Chao,LOY C C,HE Kaiming,et al. Image super–resolution using deep convolutional networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(2):295−307.

[8] DONG Chao，LOY C C，TANG Xiaoou. Accelerating the super–resolution convolutional neural network[C]//Computer Vision– ECCV 2016. Cham：Springer，2016：391–407.

[9] SHI Wenzhe，CABALLERO J，HUSZÁR F，et al. Real–time single image and video super–resolution using an efficient sub–pixel convolutional neural network[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas：IEEE，2016：1874–1883.

[10] KIM J，LEE J K，LEE K M. Accurate image super–resolution using very deep convolutional networks[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas：IEEE，2016：1646–1654.

[11] LIM B，SON S，KIM H，et al. Enhanced deep residual networks for single image super–resolution[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Honolulu：IEEE，2017：1132–1140.

[12] ZHANG Yulun，LI Kunpeng，LI Kai，et al. Image super–resolution using very deep residual channel attention networks[C]//The European Conference on Computer Vision (ECCV). Cham：Springer，2018：286–301.

[13] GOODFELLOW I,POUGET–ABADIE J,MIRZA M,et al. Generative adversarial networks[J]. Communications of the ACM,2020,63(11):139−144.

[14] LEDIG C，THEIS L，HUSZÁR F，et al. Photo–realistic single image super–resolution using a generative adversarial network[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu：IEEE，2017：105–114.

[15] TREVIZAN W A,JESUS C M D. Application of GAN to resolution enhancement of LWD real–time image logs to support decision making[J]. Petrophysics,2023,64(6):890−899.

[16] RONG Yi,JIA Mingbin,ZHAN Yufei,et al. SR–RDFAN–LOG:Arbitrary–scale logging image super–resolution reconstruction based on residual dense feature aggregation[J]. Geoenergy Science and Engineering,2024,240:213042.

[17] KANG Zhengming,LI Xin,NI Weining,et al. Using logging while drilling resistivity imaging data to quantitatively evaluate fracture aperture based on numerical simulation[J]. Journal of Geophysics and Engineering,2021,18(3):317−327.

[18] LI Hu,ZHU Jun,XIONG Yanchun,et al. On the depth of detection of logging–while–drilling resistivity measurements for looking–around and looking–ahead applications[J]. Interpretation,2020,8(3):SL151−SL158.

[19] WANG Lei,DENG Shaogui,ZHANG Pan,et al. Detection performance and inversion processing of logging–while–drilling extra–deep azimuthal resistivity measurements[J]. Petroleum Science,2019,16(5):1015−1027.

[20] 康正明,柯式镇,李新,等. 钻头电阻率测井仪器探测特性研究[J]. 石油科学通报,2017,2(4):457−465

KANG Zhengming,KE Shizhen,LI Xin,et al. The detection characteristics study of the at–bit resistivity logging tool[J]. Petroleum Science Bulletin,2017,2(4):457−465

[21] AO Yile,TIAN Fei,LU Wenkai,et al. Improving logging–while–drilling azimuthal imaging with deep learning super–resolution[J]. IEEE Transactions on Geoscience and Remote Sensing,2025,63:4500415.

[22] ALQAHTANI H,KAVAKLI–THORNE M,KUMAR G. Applications of generative adversarial networks (GANs):An updated review[J]. Archives of Computational Methods in Engineering,2021,28(2):525−552.

[23] SAXENA D,CAO Jiannong. Generative adversarial networks (GANs):Challenges,solutions,and future directions[J]. ACM Computing Surveys,2021,54(3):1−42.

[24] 乔天骄,陈昱丞,李子润,等. 基于注意力机制改进卷积神经网络的负荷预测研究[J]. 自动化与仪器仪表,2025(7):273−277

QIAO Tianjiao,CHEN Yucheng,LI Zirun,et al. Research on improving convolutional neural network load forecasting based on attention mechanism[J]. Automation & Instrumentation,2025(7):273−277

[25] AHMAD Z,JAFFRI Z U A,CHEN Meng,et al. Understanding GANs:Fundamentals,variants,training challenges,applications,and open problems[J]. Multimedia Tools and Applications,2025,84(12):10347−10423.

[26] SALIMANS T,GOODFELLOW I,ZAREMBA W,et al. Improved techniques for training GANs[J]. Advances in Neural Information Processing Systems,2016,29:03498.

[27] LI Huan. Image super–resolution algorithm based on RRDB model[J]. IEEE Access,2021,9:156260−156273.

[28] MANIATOPOULOS A,MITIANOUDIS N. Learnable leaky ReLU (LeLeLU):An alternative accuracy–optimized activation function[J]. Information,2021,12(12):513.

[29] 康正明,武辰升,杨国栋,等. 生成对抗网络算法在电成像测井裂缝空白条带填充中的应用[J]. 煤田地质与勘探,2025,53(3):197−207

KANG Zhengming,WU Chensheng,YANG Guodong,et al. Application of a generative adversarial network algorithm to filling blank strips of fractures in formation microresistivity imaging images[J]. Coal Geology & Exploration,2025,53(3):197−207

[30] HAN Xiaoyi,MA Xiaochuan,LI Houpu,et al. A global–information–constrained deep learning network for digital elevation model super–resolution[J]. Remote Sensing,2023,15(2):305.

[31] SHI Fang,LIAO Hualin,QU Fengtao,et al. Collaborative–driven reservoir formation pressure prediction using GAN–ML models and well logging data[J]. Geoenergy Science and Engineering,2024,242:213271.

[32] FAN Jianbao,ZHANG Wenxiu,CHEN Wenxuan,et al. Inversion based on deep learning of logging–while–drilling directional resistivity measurements[J]. Journal of Petroleum Science and Engineering,2022,208:109677.

[33] CHICCO D,WARRENS M J,JURMAN G. The coefficient of determination R–squared is more informative than SMAPE,MAE,MAPE,MSE and RMSE in regression analysis evaluation[J]. PeerJ. Computer Science,2021,7:e623.

[34] KINGMA D P，BA J. Adam：A method for stochastic optimization[C]//The 3^rd International Conference for Learning Representations (ICLR). San Diego，2015：267–277.