SaltISCG: Interactive Salt Segmentation Method Based on CNN and Graph Cut

Zhang, Hao; Zhu, Peimin; Liao, Zhiying; Li, Zewei

doi:10.1109/tgrs.2022.3172997

Cited by 13 publications

(4 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to apply the user's interactions to salt recognition, we refer to the 2D interaction scheme in [38] to convert the interaction points into 3D Euclidean distance maps (3D EDMs). Assuming that the coordinates of the interaction point in the 3D space are ( , , ) l m n , the 3D EDM of the point is calculated as…”

Section: Three Dimensional Euclidean Distance Mapsmentioning

confidence: 99%

“…We use a 3D U-net architecture to construct a CNN model for 3D salt identification, as shown in Figure 4. In order to make the model suitable for 3D data, we use 3D operators to replace the convolutional layer, pooling layer, and upsampling layer in the 2D U-net model [38]. Since there is a positive correlation between the receptive field in the 3D U-net model and the size of the input data, a larger receptive field can enable the model to extract global information from seismic data.…”

Section: Three-dimensional Network Architecturementioning

confidence: 99%

“…The network takes, as the combined input, the previous mask output along with a seismic image in a new moving window to predict the salt in this window. Zhang et al [38] propose combining a CNN and a graph-cut to interactively extract salt in 2D seismic profiles. However, these methods still have the following problems: (1) Due to the small sample size of the training set, the results predicted by the pre-trained models are defective, which affects the final recognition result.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SaltISNet3D: Interactive Salt Segmentation from 3D Seismic Images Using Deep Learning

2023

View full text Add to dashboard Cite

Salt interpretation using seismic data is essential for structural interpretation and oil and gas exploration. Although deep learning has made great progress in automatic salt image segmentation, it is often difficult to obtain satisfactory results in complex situations. Thus, interactive segmentation with human intervention can effectively replace the fully automatic method. However, the current interactive segmentation cannot be directly applied to 3D seismic data and requires a lot of human interaction. Because it is difficult to collect 3D seismic data containing salt, we propose a workflow to simulate salt data and use a large amount of 3D synthetic salt data for training and testing. We use a 3D U-net model with skip connections to improve the accuracy and efficiency of salt interpretation. This model takes 3D seismic data volume with a specific size as an input and generates a salt probability volume of the same size as an output. To obtain more detailed salt results, we utilize a 3D graph-cut to ameliorate the results predicted by the 3D U-net model. The experimental results indicate that our method can achieve more efficient and accurate segmentation of 3D salt bodies than fully automatic methods.

show abstract

Section: Three Dimensional Euclidean Distance Mapsmentioning

confidence: 99%

Section: Three-dimensional Network Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SaltISNet3D: Interactive Salt Segmentation from 3D Seismic Images Using Deep Learning

2023

View full text Add to dashboard Cite

show abstract

“…In recent years, deep learning technology represented by deep neural network (DNN) has been widely used in seismic interpretation, such as fault detection [16] [35] [29], saltbody delineation [11] [40], and identification of geological structural elements [13]. DNN outperform traditional machine learning methods by a large margin in many domains.…”

mentioning

confidence: 99%

Contrastive Learning Approach for Semi-Supervised Seismic Facies Identification Using High-Confidence Representations

Li¹,

Liu²,

Dou³

et al. 2022

Preprint

View full text Add to dashboard Cite

The manual seismic facies annotation relies heavily on the experience of seismic interpreters, and the distribution of seismic facies in adjacent locations is very similar, which means that much of the labeling is costly repetitive work. However, we found that training the model with only a few evenly sampled labeled slices still suffers from severe classification confusion, that is, misidentifying one class of seismic facies as another. To address this issue, we propose a semi-supervised seismic facies identification method using features from unlabeled data for contrastive learning. We sample features in regions with high identification confidence, and use a pixel-level instance discrimination task to narrow the intra-class distance and increase the inter-class distance. Instance discrimination encourages the latent space to produce more distinguishable decision boundaries and reduces the bias in the features of the same class. Our method only needs to extend one branch to compute the contrastive loss without extensive changes to the network structure. We have conducted experiments on two public seismic surveys, SEAM AI and Netherlands F3, and the proposed model achieves an IOU score of more than 90 using only 1% of the annotations in the F3 survey. We have made our codes and pre-trained models publicly available: www.github.com/upcliuwenlong/CONSS.

show abstract

Density Map-based vehicle counting in remote sensing images with limited resolution

Guo

et al. 2022

ISPRS Journal of Photogrammetry and Remote Sensing

View full text Add to dashboard Cite

SaltISCG: Interactive Salt Segmentation Method Based on CNN and Graph Cut

Cited by 13 publications

References 52 publications

SaltISNet3D: Interactive Salt Segmentation from 3D Seismic Images Using Deep Learning

SaltISNet3D: Interactive Salt Segmentation from 3D Seismic Images Using Deep Learning

Contrastive Learning Approach for Semi-Supervised Seismic Facies Identification Using High-Confidence Representations

Density Map-based vehicle counting in remote sensing images with limited resolution

Contact Info

Product

Resources

About