EUVIMEDEUVIMED entdecken
Crossref

U-Net Based Salt Segmentation Using Deep Learning for Seismic Image Analysis

Bindushree S and Dr. Kruthi R

International Journal of Advanced Research in Science Communication and Technology · 29.07.2026 · 726

DOI: 10.48175/ijarsct-37880 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 29.07.2026. Verfasst von Bindushree S and Dr. Kruthi R. Die Arbeit nennt 5 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate salt segmentation from seismic images plays a vital role in petroleum exploration and geological interpretation. This paper presents a U-Net Based Salt Segmentation system that automatically detects and segments salt regions using deep learning. The proposed model is implemented using the PyTorch framework and trained on the TGS Salt Identification Challenge Dataset. The U-Net architecture performs pixel-level semantic segmentation through an encoder-decoder network with skip connections, enabling precise identification of salt structures. In addition to segmentation, the system provides quantitative analysis, including salt coverage, salt body count, prediction confidence, and uncertainty estimation. It also generates visualization outputs such as the predicted mask, overlay image, probability heatmap, uncertainty map, and three-dimensional visualization of segmented salt regions. An automatic PDF report generation module summarizes the prediction results and analytical information into a structured report. Experimental results demonstrate that the proposed system provides accurate, reliable, and efficient salt segmentation, reducing manual interpretation effort and supporting improved geological decision-making

Zitieren

APA

R, B. S. A. D. K. (2026). U-Net Based Salt Segmentation Using Deep Learning for Seismic Image Analysis. International Journal of Advanced Research in Science Communication and Technology, 726. https://doi.org/10.48175/ijarsct-37880

Vancouver

R BSADK. U-Net Based Salt Segmentation Using Deep Learning for Seismic Image Analysis. International Journal of Advanced Research in Science Communication and Technology. 2026:726. doi:10.48175/ijarsct-37880

BibTeX
@article{R2026,
  author = {R, Bindushree S and Dr. Kruthi},
  title = {U-Net Based Salt Segmentation Using Deep Learning for Seismic Image Analysis},
  journal = {International Journal of Advanced Research in Science Communication and Technology},
  year = {2026},
  pages = {726},
  doi = {10.48175/ijarsct-37880}
}
Kostenlos starten →