Extraction of strong beadlike reflections for a carbonate-karst reservoir using a tensor-based adaptive mathematical morphology

Wang, Yi; Lu, Wenkai; Wang, Benfeng; Chen, Jianjun; Ding, Jianfeng

doi:10.1088/1742-2140/aa76d0

Cited by 2 publications

(2 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These reservoir identification methods are less accurate for identifying caves. Generally, the dissolution caves appear as string of beadlike reflections (SBRs) on the seismic migration profiles, which are considered as an indicator for highquality reservoirs (Wang et al, 2017). Correct detection and extraction of SBRs from seismic imaging profiles are helpful for quantitative description of karst cave reservoirs.…”

Section: Introductionmentioning

confidence: 99%

“…Correct detection and extraction of SBRs from seismic imaging profiles are helpful for quantitative description of karst cave reservoirs. Wang et al (2017) propose a tensor-based adaptive mathematical morphology to extract the SBRs from seismic images and remove the effects of seismic events and noise. These conventional methods cannot cleanly extract the SBRs from seismic images nor accurately depict the karst cave positions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic karst cave detection from seismic images via a convolutional neural network and transfer learning

Huang

Huang²,

Ma³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

The identification of karst caves in seismic imaging profiles is a key step for reservoir interpretation, especially for carbonate reservoirs with extensive cavities. In traditional methods, karst caves are usually detected by looking for the string of beadlike reflections (SBRs) in seismic images, which are extremely time-consuming and highly subjective. We propose an end-to-end convolutional neural network (CNN) to automatically and effectively detect karst caves from 2D seismic images. The identification of karst caves is considered as an image recognition problem of labeling a 2D seismic image with ones on caves and zeros elsewhere. The synthetic training data set including the seismic imaging profiles and corresponding labels of karst caves are automatically generated through our self-defined modeling and data augmentation method. Considering the extreme imbalance between the caves (ones) and non-caves (zeros) in the labels, we adopt a class-balanced loss function to maintain good convergence during the training process. The synthetic tests demonstrate the capability and stability of our proposed network, which is capable of detecting the karst caves from the seismic images contaminated with severe random noise. The physical simulation data example also confirms the effectiveness of our method. To overcome the generalization problem of training the neural network with only synthetic data, we introduce the transfer learning strategy and obtain good results on the seismic images of the field data.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%