Convolutional neural networks for automated seismic interpretation

Waldeland, Anders U.; Jensen, Are Charles; Gelius, Leiv‐J.; Solberg, Anne

doi:10.1190/tle37070529.1

Cited by 166 publications

(77 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Motivated by good results in many areas, CNNs are becoming researchers' default choice for the segmentation of seismic images and identification of salt deposits. A huge number of papers [34][35][36][37][38][39][40][41] in 2018 and 2019 supports the claim. Dramsch and Lüthje [34] evaluated several classification deep CNNs with transfer learning to identify nine different seismic textures from 65 × 65 pixel patches.…”

Section: Related Workmentioning

confidence: 89%

Identification of Salt Deposits on Seismic Images Using Deep Learning Method for Semantic Segmentation

Milosavljević

2020

IJGI

View full text Add to dashboard Cite

Several areas of Earth that are rich in oil and natural gas also have huge deposits of salt below the surface. Because of this connection, knowing precise locations of large salt deposits is extremely important to companies involved in oil and gas exploration. To locate salt bodies, professional seismic imaging is needed. These images are analyzed by human experts which leads to very subjective and highly variable renderings. To motivate automation and increase the accuracy of this process, TGS-NOPEC Geophysical Company (TGS) has sponsored a Kaggle competition that was held in the second half of 2018. The competition was very popular, gathering 3221 individuals and teams. Data for the competition included a training set of 4000 seismic image patches and corresponding segmentation masks. The test set contained 18,000 seismic image patches used for evaluation (all images are 101 × 101 pixels). Depth information of the sample location was also provided for every seismic image patch. The method presented in this paper is based on the author’s participation and it relies on training a deep convolutional neural network (CNN) for semantic segmentation. The architecture of the proposed network is inspired by the U-Net model in combination with ResNet and DenseNet architectures. To better comprehend the properties of the proposed architecture, a series of experiments were conducted applying standardized approaches using the same training framework. The results showed that the proposed architecture is comparable and, in most cases, better than these segmentation models.

show abstract

Section: Related Workmentioning

confidence: 89%

Identification of Salt Deposits on Seismic Images Using Deep Learning Method for Semantic Segmentation

Milosavljević

2020

IJGI

View full text Add to dashboard Cite

show abstract

“…Waldeland et al . () demonstrated how CNNs could be used to classify different seismic textures with special emphasis on salt bodies. Qian et al .…”

Section: Introductionmentioning

confidence: 99%

“…Rentsch et al (2014) trained a neural network to identify SI noise for towed-streamer acquisitions. Waldeland et al (2018) demonstrated how CNNs could be used to classify different seismic textures with special emphasis on salt bodies. Qian et al (2018) applied a deep convolutional autoencoder (DCAE) network in seismic facies recognition based on prestack seismic data.…”

mentioning

confidence: 99%

Attenuation of marine seismic interference noise employing a customized U‐Net

Sun

Slang

Elboth

et al. 2020

Geophysical Prospecting

View full text Add to dashboard Cite

A B S T R A C TMarine seismic interference noise occurs when energy from nearby marine seismic source vessels is recorded during a seismic survey. Such noise tends to be well preserved over large distances and causes coherent artefacts in the recorded data. Over the years, the industry has developed various denoising techniques for seismic interference removal, but although well performing, they are still time-consuming in use. Machine-learning-based processing represents an alternative approach, which may significantly improve the computational efficiency. In the case of conventional images, autoencoders are frequently employed for denoising purposes. However, due to the special characteristics of seismic data as well as the noise, autoencoders failed in the case of marine seismic interference noise. We, therefore, propose the use of a customized U-Net design with element-wise summation as part of the skip-connection blocks to handle the vanishing gradient problem and to ensure information fusion between high-and low-level features. To secure a realistic study, only seismic field data were employed, including 25,000 training examples. The customized U-Net was found to perform well, leaving only minor residuals, except for the case when seismic interference noise comes from the side. We further demonstrate that such noise can be treated by slightly increasing the depth of our network. Although our customized U-Net does not outperform a standard commercial algorithm in quality, it can (after proper training) read and process one single shot gather in approximately 0.02 s. This is significantly faster than any existing industry denoising algorithm. In addition, the proposed network processes shot gathers in a sequential order, which is an advantage compared with industry algorithms that typically require a multi-shot input to break the coherency of the noise.

show abstract

“…Di, Wang and AlRegib ; Waldeland et al . ; Zeng, Jiang and Chen ; Shi, Wu and Fomel ). However, little work has been done towards an in‐depth comparison of these techniques in seismic saltbody delineation.…”

Section: Introductionmentioning

confidence: 99%

“…Di, Shafiq and AlRegib 2017;Huang, Dong and Clee 2017;Guitton, Wang and Guitton 2017) and saltbody delineation (e.g. Di, Wang and AlRegib 2018a;Waldeland et al 2018;Zeng, Jiang and Chen 2018;Shi, Wu and Fomel 2019). However, little work has been done towards an in-depth comparison of these techniques in seismic saltbody delineation.…”

mentioning

confidence: 99%

A comparison of seismic saltbody interpretation via neural networks at sample and pattern levels

AlRegib

2019

Geophysical Prospecting

View full text Add to dashboard Cite

Saltbodies are important subsurface structures that have significant implications for hydrocarbon accumulation and sealing in petroleum reservoirs, and accurate saltbody imaging and delineation is now greatly facilitated with the availability of three‐dimensional seismic surveying. However, with the growing demand for larger survey coverage and higher imaging resolution, the size of seismic data is increasing dramatically. Correspondingly, manual saltbody interpretation fails to offer an efficient solution, particularly in exploration areas of complicated salt intrusion history. Recently, artificial intelligence is attracting great attention from geoscientists who desire to utilize the popular machine learning technologies for evolving the interpretational tools capable of mimicking an experienced interpreter's intelligence. This study first implements two popular machine learning tools, the multi‐layer perceptron and the convolutional neural network, for delineating seismic saltbodies at sample and pattern levels, respectively, then compares their performance through applications to the synthetic SEAM seismic volume, and moreover tentatively investigates what contributes to the better convolutional neural network delineation. Specifically, the multi‐layer perceptron scheme is capable of efficiently utilizing an interpreter's knowledge by selecting, pre‐conditioning and integrating a set of seismic attributes that best highlight the target saltbodies, whereas the convolutional neural network scheme makes it possible for saltbody delineation directly from seismic amplitude and thus significantly reduces the dependency on attribute selection from interpreters. It is concluded that the better performance from the convolutional neural network scheme results from two factors. First, the convolutional neural network builds the mapping relationship between the seismic signals and the saltbodies using the original seismic amplitude instead of manually selected seismic attributes, so that the negative impact of using less representative attributes is virtually eliminated. Second and more importantly, the convolutional neural network defines, learns and identifies the saltbodies by utilizing local seismic reflection patterns, so that the seismic noises and processing artefacts of distinct patterns are effectively identified and excluded.

show abstract

Convolutional neural networks for automated seismic interpretation

Cited by 166 publications

References 13 publications

Identification of Salt Deposits on Seismic Images Using Deep Learning Method for Semantic Segmentation

Identification of Salt Deposits on Seismic Images Using Deep Learning Method for Semantic Segmentation

Attenuation of marine seismic interference noise employing a customized U‐Net

A comparison of seismic saltbody interpretation via neural networks at sample and pattern levels

Contact Info

Product

Resources

About