.png){kind=icon}
.png){kind=icon}
Return
Horizon Picking on Subbottom Profiles Using Wavelet Multiresolution Analysis
.png){kind=icon}
PRE
AI
DOI：10.1109/JOE.2025.3617912.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Horizon picking underpins the interpretation of subbottom profile (SBP) data, acoustical inversion, and related applications. We present a horizon-picking method based on an improved wavelet multiresolution analysis. The approach uses low-resolution images from multiscale decomposition to track lateral layer continuity and combines nonmaximum suppression with hysteresis thresholding to achieve uninterrupted detection. A graph-based linking algorithm further refines edges and improves accuracy. To correct horizontal offsets and localization errors caused by phase distortion during wavelet reconstruction, we introduce a morphology-guided reconstruction strategy. Key parameters are determined in synthetic experiments and validated on field data. Compared with traditional approaches, the proposed method yields more continuous, accurate, and high-resolution horizons. The method’s effectiveness is demonstrated on SBP data acquired in Jiaozhou Bay.
Keywords:
Edge detection
horizon picking
multiresolution analysis (MRA)
subbottom profiles (SBP)
wavelet transform
Journal
IF:
5.3
Papers:
2.6K
Citations:
7.4K