岩性智能识别技术是优化油气勘探决策的核心环节,其关键在于快速解译录井岩屑图像所蕴含的地质信息。随着录井技术的发展,岩屑图像数量剧增且复杂性增加,急需一种快速且准确的岩性识别方法。本文提出了一种基于改进EfficientNet的深度学习与迁移学习模型,用于8种岩屑的分类识别。与ResNet、Inception等经典神经网络方法相比,EfficientNet的迁移学习模型的泛化能力更强,计算速度更快,识别测试集准确率能达到97.84%。
Intelligent lithology identification technology serves as a critical component in optimizing oil and gas exploration decision-making, with its core focus lying in the rapid interpretation of geological information embedded in mud logging cuttings images. As mud logging technology advances, the volume of obtainable cuttings images has surged dramatically while the complexity and ambiguity of cuttings have intensified, creating an urgent demand for rapid and accurate lithology identification methods in field operations. This study develops an improved EfficientNet-based deep learning and transfer learning model to classify eight common types of cuttings. Compared to classical neural networks like ResNet and Inception, the EfficientNet-based transfer learning model shows better generalization and faster computation, achieving a recognition accuracy of 97.84% on the test dataset.
[1] 王新领, 祝新益, 张宏兵, 等. 基于随机树嵌入的随钻测井岩性识别方法[J]. 吉林大学学报(地球科学版), 2024, 54(2): 701-708. DOI: 10.13278/j.cnki.jjuese.20230193.
[2] 张众, 郭金涛, 李鹏. 录井技术对钻井井控中的作用分析[J]. 石化技术, 2025, 32(1): 291-293. DOI: 10.3969/j.issn.1006-0235.2025.01.101.
[3] 潘柯宇, 卢正东, 唐明, 等. 四川盆地SYS构造录井岩屑识别方法探讨及应用[J]. 钻采工艺, 2020, 43(z1): 24-26,34. DOI: 10.3969/J.ISSN.1006-768X.2020.增.07.
[4] 杨宇轩, 钟宝荣. 基于改进ConvNeXt模型的轻量化岩屑岩性识别方法[J]. 现代电子技术, 2024, 47(24): 131-136. DOI: 10.16652/j.issn.1004-373x.2024.24.020.
[5] 马世拓, 班一杰, 戴陈至力. 卷积神经网络综述[J]. 现代信息科技, 2021, 5(2): 11-15. DOI: 10.19850/j.cnki.2096-4706.2021.02.003.
[6] Karen Simonyan, Andrew Zisserman. Very Deep Convolutional Networks For Large-scale Image Recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
[7] He K M, Zhang X Y, Ren S Q, et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016: 770-778. DOI: 10.1109/CVPR.2016.90.
[8] Szegedy C, Liu W, Jia Y Q, et al.Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 7-12, 2015, Boston, MA, USA. IEEE, 2015: 1-9. DOI: 10.1109/CVPR.2015.7298594.
[9] Xu Z H, Ma W, Lin P, et al.Deep learning of rock microscopic images for intelligent lithology identification: Neural network comparison and selection[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2022, 14(4): 1140-1152. DOI: 10.1016/j.jrmge.2022.05.009.
[10] 杨磊, 熊昶, 刘文超, 等. 基于改进ResNet深度残差网络的岩屑岩性识别研究[J]. 长江大学学报(自然科学版), 2023, 20(2): 11-19. DOI: 10.3969/j.issn.1673-1409.2023.02.002.
[11] Liu Z F, Zhang J J, Li Y G, et al. Lithology prediction of one-dimensional residual network based on regularization constraints[J]. Journal of Petroleum Science and Engineering, 2022, 215, Part B: 110620. DOI: 10.1016/j.petrol.2022.110620.
[12] 凌同华, 陈梓浓, 张胜, 等. 岩石岩性轻量化网络识别方法及其隧道工程应用[J]. 铁道科学与工程学报,2023, 20(9): 3604-3615. DOI: 10.19713/j.cnki.43-1423/u.T20221939.
[13] 董文豪, 张怀. 基于迁移学习的岩屑岩性识别[J]. 中国科学院大学学报, 2023, 40(6): 743-750.
[14] Liu C Z, Li M C, Zhang Y, et al.An enhanced rock mineral recognition method integrating a deep learning model and clustering algorithm[J]. Minerals, 2019, 9(9): 516. DOI: 10.3390/min9090516.
[15] Wang H J.Intelligent identification of logging cuttings based on deep learning[J]. Energy Reports, 2022, 8: 1-7. DOI: 10.1016/j.egyr.2022.10.049.
[16] Tan M, Le Q V.(2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. In International Conference on Machine Learning (ICML). arXiv preprint arXiv:1905.11946.
[17] 王潇怡,罗运华,喻忠军,等. 基于EfficientNet的多通道雷达目标微动特征分类方法[J]. 系统工程与电子技术, 2024, 46(9): 2988-2998. DOI: 10.12305/j.issn.1001-506X.2024.09.11.
[18] 付军军, 陈林, 童华敏, 等. 基于改进EfficientNet的海上风机叶片早期缺陷检测及分类[J]. 电子测量技术, 2024, 47(16): 110-119. DOI: 10.19651/j.cnki.emt.2416268.
[19] 沈先耿. 深度学习综述[J]. 数字化用户,2017(11):63. DOI: 10.3969/j.issn.1009-0843.2017.11.057.
[20] 马世龙, 乌尼日其其格, 李小平. 大数据与深度学习综述[J]. 智能系统学报, 2016, 11(6): 728-742. DOI: 10.11992/tis.201611021.
[21] 刘嘉雄, 周骏. 基于深度学习的风格迁移方法综述[J]. 人工智能科学与工程, 2024, 2(1): 1-19. DOI: 10.3969/j.issn.1002-6339.2025.02.014
[22] 黄兆培, 张峰源, 赵金明, 等. 情感识别中的迁移学习问题综述[J]. 信号处理, 2023, 39(4): 588-615. DOI: 10.16798/j.issn.1003-0530.2023.04.002.
[23] 华晨晨, 王奭寅, 徐茂森, 等. 基于改进VGG16的射流式鱼泵内鱼类损伤的图像识别与分类[J]. 中国海洋大学学报(自然科学版), 2024, 54(11): 49-57. DOI: 10.16441/j.cnki.hdxb.20230216.
[24] 于营, 杨婷婷, 杨博雄. 混淆矩阵分类性能评价及Python实现[J]. 现代计算机, 2021(20): 70-73, 79. DOI: 10.3969/j.issn.1007-1423.2021.20.014.
[25] 邹永康, 范琴, 王佩瑜. 改进UNet++与联合损失函数的建筑物提取方法[J]. 长江信息通信, 2024, 37(6): 78-80. DOI: 10.20153/j.issn.2096-9759.2024.06.023.
[26] 阚涛, 高哲, 杨闯. 采用分数阶动量的卷积神经网络随机梯度下降法[J]. 模式识别与人工智能, 2020, 33(6): 559-567. DOI: 10.16451/j.cnki.issn1003-6059.202006009.
