Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
An Efficient Approach for Semantic Segmentation of Salt Domes in Seismic Images Using Improved UNET Architecture
https://orcid.org/http://orcid.org/0000-0001-6765-9342
Many areas of Earth’s surface with large accumulations of gas and oil even have huge deposits of salt under the surface. Exploring such deposits helps many countries to increase the storage capacity of their Petroleum reserves and explore new ones. But finding such deposits is a herculean task. Expert seismic imaging requires human interpretation of salt bodies. But this leads to very biased and highly variable translations. So the idea behind this paper is to build an approach that accurately and automatically identifies if the seismic image contains any region of salt deposit or not. If a surface is found to have salt deposits, then it may contain the accumulations of oil or gas and even the salt domes or caverns can be used as a storage site for already available petroleum or oil. Since semantic segmentation classifies every pixel in the given image to its class label, this can be used to segment the salt deposits from the provided seismic images. In this paper, we introduce a variation of UNet, a popular segmentation model, for seismic image segmentation. We have added a batch normalization layer following every convolution layer as a deeper network helps extract better features which turned out to be true. Some interesting findings of this work are that the augmentation works well as this avoids over-fitting of the model. Sharpening as a post-processing technique has come up with a considerable amount of rise in the performance. Here, we prefer to use the metric as Intersection over Union (IoU) instead of accuracy as it is not as affected by the class imbalances that are inherent in foreground/background segmentation tasks. With the proposed methodology, we achieve an averaged IoU of 85.6 which is far better compared to the IoU achieved with the Segnet approach which stands at 77.
An Efficient Approach for Semantic Segmentation of Salt Domes in Seismic Images Using Improved UNET Architecture
https://orcid.org/http://orcid.org/0000-0001-6765-9342
Many areas of Earth’s surface with large accumulations of gas and oil even have huge deposits of salt under the surface. Exploring such deposits helps many countries to increase the storage capacity of their Petroleum reserves and explore new ones. But finding such deposits is a herculean task. Expert seismic imaging requires human interpretation of salt bodies. But this leads to very biased and highly variable translations. So the idea behind this paper is to build an approach that accurately and automatically identifies if the seismic image contains any region of salt deposit or not. If a surface is found to have salt deposits, then it may contain the accumulations of oil or gas and even the salt domes or caverns can be used as a storage site for already available petroleum or oil. Since semantic segmentation classifies every pixel in the given image to its class label, this can be used to segment the salt deposits from the provided seismic images. In this paper, we introduce a variation of UNet, a popular segmentation model, for seismic image segmentation. We have added a batch normalization layer following every convolution layer as a deeper network helps extract better features which turned out to be true. Some interesting findings of this work are that the augmentation works well as this avoids over-fitting of the model. Sharpening as a post-processing technique has come up with a considerable amount of rise in the performance. Here, we prefer to use the metric as Intersection over Union (IoU) instead of accuracy as it is not as affected by the class imbalances that are inherent in foreground/background segmentation tasks. With the proposed methodology, we achieve an averaged IoU of 85.6 which is far better compared to the IoU achieved with the Segnet approach which stands at 77.
An Efficient Approach for Semantic Segmentation of Salt Domes in Seismic Images Using Improved UNET Architecture
https://orcid.org/http://orcid.org/0000-0001-6765-9342
Many areas of Earth’s surface with large accumulations of gas and oil even have huge deposits of salt under the surface. Exploring such deposits helps many countries to increase the storage capacity of their Petroleum reserves and explore new ones. But finding such deposits is a herculean task. Expert seismic imaging requires human interpretation of salt bodies. But this leads to very biased and highly variable translations. So the idea behind this paper is to build an approach that accurately and automatically identifies if the seismic image contains any region of salt deposit or not. If a surface is found to have salt deposits, then it may contain the accumulations of oil or gas and even the salt domes or caverns can be used as a storage site for already available petroleum or oil. Since semantic segmentation classifies every pixel in the given image to its class label, this can be used to segment the salt deposits from the provided seismic images. In this paper, we introduce a variation of UNet, a popular segmentation model, for seismic image segmentation. We have added a batch normalization layer following every convolution layer as a deeper network helps extract better features which turned out to be true. Some interesting findings of this work are that the augmentation works well as this avoids over-fitting of the model. Sharpening as a post-processing technique has come up with a considerable amount of rise in the performance. Here, we prefer to use the metric as Intersection over Union (IoU) instead of accuracy as it is not as affected by the class imbalances that are inherent in foreground/background segmentation tasks. With the proposed methodology, we achieve an averaged IoU of 85.6 which is far better compared to the IoU achieved with the Segnet approach which stands at 77.
An Efficient Approach for Semantic Segmentation of Salt Domes in Seismic Images Using Improved UNET Architecture
J. Inst. Eng. India Ser. B
Bodapati, Jyostna Devi (Autor:in) / Sajja, RamaKrishna (Autor:in) / Naralasetti, Veeranjaneyulu (Autor:in)
Journal of The Institution of Engineers (India): Series B ; 104 ; 569-578
01.06.2023
10 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
| DOAJ | 2021
| Engineering Index Backfile | 1929
Detail-sensitive 3D-UNet for pulmonary airway segmentation from CT images
| Springer Verlag | 2024
Seismic vulnerability of Iranian historical domes
| British Library Conference Proceedings | 2003
Seismic Behavior of Single Layer Schwedler Domes
| Online Contents | 2014