[1]张利霞,高俊涛,马 强,等.基于改进 UNet++的地震断层识别方法研究[J].计算机技术与发展,2023,33(08):199-205.[doi:10. 3969 / j. issn. 1673-629X. 2023. 08. 029]
　ZHANG Li-xia,GAO Jun-tao,MA Qiang,et al.Research on Seismic Fault Identification Methods Based on Improved UNet++[J].,2023,33(08):199-205.[doi:10. 3969 / j. issn. 1673-629X. 2023. 08. 029]

点击复制

基于改进 UNet++的地震断层识别方法研究()

分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

33

期数:

2023年08期

页码:

199-205

栏目:

新型计算应用系统

出版日期:

2023-08-10

文章信息/Info

Title:

Research on Seismic Fault Identification Methods Based on Improved UNet++

文章编号:

1673-629X(2023)08-0199-07

作者:

张利霞¹ ; 高俊涛¹ ; 马 强² ; 杨润湉¹ ; 王志宝¹ ; 李 菲²

1. 东北石油大学 计算机与信息技术学院,黑龙江 大庆 163318;

2. 黑龙江八一农垦大学 信息与电气工程学院,黑龙江 大庆 163319

Author(s):

ZHANG Li-xia1 ; GAO Jun-tao1 ; MA Qiang2 ; YANG Run-tian1 ; WANG Zhi-bao1 ; LI Fei2

1. School of Computer and Information Technology,Northeast Petroleum University,Daqing 163318,China;

2. School of Information and Electrical Engineering,Heilongjiang Bayi Agricultural University,Daqing 163319,China

关键词:

地震断层识别; 图像分割; UNet++模型; CBAM 注意力模块; DropBlock

Keywords:

seismic fault identification; image segmentation; UNet++ model; CBAM attention module; DropBlock

分类号:

TP18

DOI:

10. 3969 / j. issn. 1673-629X. 2023. 08. 029

摘要:

断层解释是油气勘探开发过程中的重要工作,但是随着勘探规模的增大,传统的人工解释断层的方法已经无法满足实际生产的需要。 针对人工标注断层特征费时费力、传统断层识别结果连续性不足的局限,以及地震资料中断层与非断层样本分类不均衡的问题,提出基于 CBAM-UNet++模型的地震断层识别方法。 采用合成地震数据自动生成地震数据和断层标签,提高断层标注的效率。 首先,将 CBAM 注意力模块引入 UNet++,从通道和空间两个维度抑制地震振幅信号干扰,增强地震断层的检测能力,采用 DropBlock 模块抑制网络中产生的过拟合问题;其次,引入 Dice Loss 损失函数用于减小断层识别任务中数据不均衡问题对模型的影响;再次,对断层预测结果进行霍夫变换,提取骨架,使断层预测结果更好地应用于地质目标;最后,在合成地震数据集、北海地区 F3 区块真实地震数据上评估 CBAM - UNet + + 模型,与 UNet + +、UNet、SegNet 进行对比。 结果表明,基于 CBAM-UNet++的断层识别方法在准确率、断层连续性方面表现优异,可自动、有效地识别地震图像中的断层。

Abstract:

Fault interpretation is an important task in the exploration and development process of oil and gas. However,as the explorationscale increases,the?

traditional manual fault interpretation can no longer satisfy the needs of actual production. In light of the limitations ofmanual fault annotation,such?

as being time-consuming & laborious,insufficient continuity of traditional fault identification results,andthe unbalanced classification of fault and non-

fault samples in seismic data,a seismic fault identification method based on CBAM-UNet++model is proposed,which adopts synthesized seismic data to automatically generate seismic data and fault labels,so as to improve the efficiency of fault annotation. Firstly,we introduce the CBAM attention module?

into UNet + +,which suppresses the signal interference ofseismic amplitude from the two dimensions of channel and space to enhance the detectability?

of seismic fault,and suppresses the overfitting in the network with the use of DropBlock module. Secondly, introduce the Dice Loss function to reduce the impact of dataimbalance on the model in fault identification task. Thirdly, perform Hough transform to the fault prediction results for skeletonextracting,so?

that the fault prediction results can be better applied to geological objectives. Finally,evaluate the CBAM-UNet++ modelbased on the synthesized seismic data set and the real seismic data of F3 block in the North Sea,and compare with UNet+ +,UNet andSegNet. The results indicate that the fault identification method based on CBAM-UNet++ model has excellent performance in terms ofaccuracy and fault continuity,and can identify faults in seismic images automatically and effectively.

常用功能