Research on fracture segmentation of FMI logging images based on deep learning
Received date: 2024-03-25
Online published: 2025-03-13
Copyright
In the process of reservoir drilling and development, it is of great significance to accurately extract, identify and evaluate the fractures in the formation to guide the drilling and development of oil and gas exploration. To solve the problem of imprecise fracture region segmentation by traditional methods, a fracture segmentation method based on Formation Micro-Scanner Image based on deep learning is proposed. Firstly, F-Criminisi algorithm is used to repair the blank strip with missing pixel information in the original FMI logging image. Then, a generative adversus-network based on U-Net is constructed, and dual attention mechanism is introduced to construct a fracture segmentation model to achieve accurate fracture segmentation under complex background. Combining pixel and edge information, loss function is designed to enable the model to more accurately segment the fracture and background region in the logging image and make the fracture boundary in the segmentation result clearer. In this paper, the proposed model is tested by using real FMI logging image of carbonate reservoir. The results show that the Dice coefficient of the proposed fracture segmentation method is 5% higher than that of the classical fracture segmentation model U-Net. This method can accurately extract fracture information from FMI logging images, and provides a basis for subsequent quantitative calculation of fracture parameters and logging interpretation, and has good practicability.
YuFan CHEN , Yang WANG , Wei JIANG , YongSheng WANG , QingYan MEI , Xin WANG . Research on fracture segmentation of FMI logging images based on deep learning[J]. Progress in Geophysics, 2025 , 40(1) : 143 -154 . DOI: 10.6038/pg2025HH0480
表1 不同修复算法结果的PSNR、SSIM值Table 1 PSNR and SSIM values of different inpainting algorithms |
| 图像 | 评价指标 | Criminisi算法 | Ou算法 | F-Criminisi算法 |
|---|---|---|---|---|
| 例1 | PSNR/dB | 24.4847 | 25.1091 | 26.5270 |
| SSIM | 0.9579 | 0.9620 | 0.9670 | |
| 例2 | PSNR/dB | 23.1494 | 25.5784 | 25.9807 |
| SSIM | 0.9549 | 0.9615 | 0.9641 | |
| 例3 | PSNR/dB | 19.2778 | 26.9933 | 28.0929 |
| SSIM | 0.9498 | 0.9670 | 0.9708 | |
| 平均 | PSNR/dB | 21.9772 | 22.6057 | 22.9451 |
| SSIM | 0.9417 | 0.9454 | 0.9465 |
表2 使用不同图像修复算法的模型ROC、PR、Dice系数曲线下面积(AUC)的比较Table 2 Comparison of the Area Under Curve (AUC) for ROC, PR, and Dice coefficients of the model using different image inpainting algorithms |
| 图像修复算法 | ROC | PR | Dice |
|---|---|---|---|
| 未修复 | 0.7876 | 0.5632 | 0.6181 |
| Criminisi算法 | 0.8057 | 0.6569 | 0.6623 |
| Ou算法 | 0.8306 | 0.6679 | 0.6471 |
| F-Criminisi算法 | 0.8583 | 0.7539 | 0.7360 |
表3 不同分割模型在ROC、PR和Dice系数的曲线下面积(AUC)的比较Table 3 Comparison of the Area Under Curve (AUC) for ROC, PR and Dice coefficient of different segmentation models |
| 分割模型 | ROC | PR | Dice |
|---|---|---|---|
| SegNet | 0.6949 | 0.6203 | 0.5288 |
| FCN | 0.7062 | 0.6669 | 0.5600 |
| U-Net | 0.8206 | 0.7256 | 0.6863 |
| Ours | 0.8583 | 0.7539 | 0.7360 |
表4 消融实验ROC、PR和Dice系数的曲线下面积(AUC)结果Table 4 The Area Under Curve (AUC) for ROC, PR and Dice coefficient of ablation experiment |
| 模型 | ROC | PR | Dice |
|---|---|---|---|
| Ours_base | 0.8206 | 0.7256 | 0.6863 |
| Ours_GAN | 0.8375 | 0.7263 | 0.7301 |
| Ours_attention | 0.8416 | 0.7377 | 0.7171 |
| Ours | 0.8583 | 0.7539 | 0.7360 |
感谢审稿专家提出的修改意见和编辑部的大力支持!
|
|
|
|
|
|
|
|
|
|
|
|
|
Fu J, Liu J, Tian H J, et al. 2019. Dual attention network for scene segmentation. //Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 3141-3149.
|
|
|
|
Goodfellow I J, Pouget-Abadie J, Mirza M, et al. 2014. Generative adversarial nets. //Proceedings of the 28th International Conference on Neural Information Processing Systems. Montreal: ACM, 2672-2680.
|
|
|
|
|
|
Isola P, Zhu J Y, Zhou T H, et al. 2017. Image-to-image translation with conditional adversarial networks. //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 5967-5976.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ronneberger O, Fischer P, Brox T. 2015. U-Net: convolutional networks for biomedical image segmentation. //18th International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich: Springer, 234-241.
|
|
|
|
|
|
Wu T. 2021. Research on automatic detection algorithm of fractures around wells under small sample conditions [Master's thesis](in Chinese). Chengdu: University of Electronic Science and Technology of China, doi: 10.27005/d.cnki.gdzku.2021.002477.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
吴桐. 2021. 小样本条件下井周裂缝自动检测算法研究[硕士论文]. 成都: 电子科技大学, doi: 10.27005/d.cnki.gdzku.2021.002477.
|
|
杨绪海, 张晓春. 2000. 利用声成像测井数据实现岩石裂缝特征的自动识别. 中国海上油气(地质), 14(6): 429-431.
|
|
|
|
|
|
|
|
|
/
| 〈 |
|
〉 |