Water Boundary Detection for SAR Images Using Region-Based Active Contours with Salient Edge Energy

Guo, Baojun; Cui, Jinlong; Gao, Beibei

doi:10.1007/s12524-022-01541-8

Cited by 3 publications

(1 citation statement)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image processing techniques or computer vision algorithms can thus assist identifying geological objects (Hale & Emanuel, 2003;Khayer et al, 2023). Active Contour Model (ACM), recognized as an image segmentation technique, have been widely applied in various domains such as medical imaging (Drozdzal et al, 2017;Li et al, 2011) and remote sensing (Guo et al, 2022). In the field of geophysics, Zheglova and Farquharson (2012) applied the level set inversion technique to travel time tomography.…”

mentioning

confidence: 99%

Seismic oolitic beach identification using a hybrid signed pressure force active contour model in the Northeastern Sichuan Basin, China

Li,

Huang,

Tang

et al. 2024

Preprint

View full text Add to dashboard Cite

Oolitic beach in Triassic Feixianguan formation is one of the majority reservoirs in the Northeastern Sichuan Basin. The precise delineation of the oolitic beach distribution using seismic data is important for characterizing the potential of gas reservoirs. Current identification workflow requires manually interpretation using seismic attributes, which is interpreter dependent and labor consuming. We propose an automatic oolitic beach identification method based on Signed Pressure Force Active Contour Model (SPFACM). To corporate with the seismic attributes, we modify SPFACM by adding an extra signed pressure force driven by the automatic threshold constraint and name it hybrid signed pressure force active contour model (HSPFACM). The automatic threshold value is defined by the maximum change of inner region variance, which approximates the outline of oolitic beaches, enhancing the stability and effectiveness of HSPFACM. We test the proposed method using field data in Northeastern Sichuan Basin. The study shows that the most sensitive attribute to characterize oolitic beach is the maximum peak amplitude, and the HSPFACM can effectively identify the oolitic beaches using this attribute. The well log interpretation shows that our identification result agrees with the locations of gas reservoirs. The accuracy of our method outperforms three state of art ACMs. The segmentation result shows that the oolitic beaches predominantly developed within the sea trough, and their occurrence decreases towards the platform margin, which agrees with the geological understanding.

show abstract