Seismic fault detection in real data using transfer learning from a convolutional neural network pre-trained with synthetic seismic data

Cunha, A.L.A. da; Pochet, Axelle; Lopes, Hélio; Gattass, Marcelo

doi:10.1016/j.cageo.2019.104344

Cited by 118 publications

(43 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Fig. 6, there is a significant difference in amplitude patterns if the seismic frequency difference is large (Cunha et al ., 2020), which have a great impact on fault detection. So we divide the experiment into two groups according to the frequency difference between synthetic seismic data and real seismic data.…”

Section: Resultsmentioning

confidence: 99%

“…16, we introduce the method of scaling seismic data in Cunha et al . (2020): the data‐scaling operation is completed by up‐sampling or down‐sampling the original seismic data, to ensure that in the same scale such as 128

\times

128

\times

128, the synthetic seismic data and the real seismic data are more similar from the perspective of the image. After using the scaled synthetic data as the training data, the fault detection results are shown in Fig.…”

Section: Resultsmentioning

confidence: 99%

“…Cunha et al . (2020) have successfully used the transfer learning in fault detection in this way: they use synthetic seismic data to train CNN as initial model, and then use real labelled data to continue training on the initial model. They confirmed the role of transfer learning in fault detection and achieved good results, but this method still relies on real labelled data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning from unlabelled real seismic data: Fault detection based on transfer learning

Zhou

Yao

et al. 2021

Geophysical Prospecting

View full text Add to dashboard Cite

Significant advances have been made towards fault detection using deep learning. However, the fault labelling of seismic data requires great human effort. The resulting small sample problem makes traditional deep learning methods difficult to achieve desired results. Existing research proposes to train a deep learning model with labelled synthetic seismic data to get good fault detection results. However, due to the complexity of the actual geological situation, there are inevitable differences between synthetic seismic data and real seismic data in many aspects such as seismic signal frequency, frequency of fault distribution and degree of noise disturbance, which lead to the fact that the deep learning model trained by synthetic seismic data is difficult to get good fault detection result in field data applications. We propose to use transfer learning to reduce the impact of data differences to solve this problem: part of the deep transfer learning model is used to learn fault‐related features. And the other part of the deep transfer learning model is used to mine common features between the real seismic data and the synthetic seismic data, which makes the deep transfer learning model more suitable for real seismic data. Compared with the latest research progress, our method can greatly improve the effect of fault detection without real data label, which can significantly save the cost of manual label processing.

show abstract

Section: Resultsmentioning

confidence: 99%

\times

128

\times

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning from unlabelled real seismic data: Fault detection based on transfer learning

Zhou

Yao

et al. 2021

Geophysical Prospecting

View full text Add to dashboard Cite

show abstract

“…As mentioned in many related works, [5] , [6] , [7] , [8] , a common challenge for a comprehensive evaluation of automatic fault recognition performance is the lack of large-scale open-source interpreted seismic datasets. Therefore, two countermeasures have emerged.…”

Section: Experimental Design Materials and Methodsmentioning

confidence: 99%

A gigabyte interpreted seismic dataset for automatic fault recognition

Jiu-lin

et al. 2021

Data in Brief

View full text Add to dashboard Cite

The lack of large-scale open-source expert-labelled seismic datasets is one of the barriers to applying today’s AI techniques to automatic fault recognition tasks. The dataset present in this article consists of a large number of processed seismic images and their corresponding fault annotations. The processed seismic images, which are originally from a seismic survey called Thebe Gas Field in the Exmouth Plateau of the Carnarvan Basin on the NW shelf of Australia, are represented in Python Numpy format, which can be easily adopted by various AI models and will facilitate cooperation with researchers in the field of computer science. The corresponding fault annotations were firstly manually labelled by expert interpreters of faults from seismic data in order to investigate the structural style and associated evolution of the basin. Then the fault interpretation and seismic survey are processed and collected using Petrel software and Python programs separately. This dataset can help to train, validate, and evaluate the performance of different automatic fault recognition workflow.

show abstract

“…Addressing CD as a classification problem, the model fused bi-temporal or multitemporal data into intermediate features that classified into binary or multi-class change maps. Examples of models include Deep Belief Network (DBN) [23], pre-trained CNN [24][25][26], tailored CNN [27,28]. The classification-based models had successfully identified forest cover change, Tsunami, and washed-away damage detection.…”

Section: A Change Detectionmentioning

confidence: 99%

Training Compact Change Detection Network for Remote Sensing Imagery

et al. 2021

View full text Add to dashboard Cite

Change Detection (CD) is a hot remote sensing topic where the change zones are highlighted by analyzing bi-temporal or multi-temporal images. Recently, Deep learning (DL) paved the road to implement various reliable change detection approaches that overcome traditional CD methods limitation. However, high performance DL based approaches have explosion number of parameters that demanded extensive computation and memory usage in addition to large volumes of training data. To address this issue, we proposed a teacher-student setting for remote sensing imagery change detection. To distill the knowledge from the over-parameterized Siamese teacher network, we proposed tiny student network that was trained using the obtained categorical distribution of probability from the teacher paired Softmax output at high temperature. Practical Swarm Optimization (PSO) was applied in order to optimally configure student architecture. Finally, ample experiments were conducted on LEVIR-CD dataset. Also, we introduced EGSAR-CD dataset, which contains of a large set of bi-temporal SAR images with 460 image pairs (256×256). Experiment results indicate that we can reach up to 5.4× reduction rate in number of parameters with loss of accuracy between 5% and 6% on the LEVIR-CD and EGSAR-CD datasets utilizing selfknowledge distillation.

show abstract

Seismic fault detection in real data using transfer learning from a convolutional neural network pre-trained with synthetic seismic data

Cited by 118 publications

References 20 publications

Learning from unlabelled real seismic data: Fault detection based on transfer learning

Learning from unlabelled real seismic data: Fault detection based on transfer learning

A gigabyte interpreted seismic dataset for automatic fault recognition

Training Compact Change Detection Network for Remote Sensing Imagery

Contact Info

Product

Resources

About