Research on Seafloor 3D Reconstruction Method Based on Sparse Measurement Points.

Area of Science:

• Geosciences
• Marine Geology
• Geophysical Surveying

Background:

• Seafloor 3D reconstruction is crucial for monitoring topography and deformation.
• Deep-sea environments yield low-resolution data, challenging existing reconstruction algorithms.
• Complex seafloor terrain variations demand higher accuracy and detail in 3D models.

Purpose of the Study:

• To develop a high-resolution 3D seafloor reconstruction method.
• To address limitations of current algorithms in handling sparse and low-resolution data.
• To enhance the accuracy and detail representation of seafloor terrain models.

Main Methods:

• A hybrid "fractal-Gaussian process" model was proposed.
• Fractal self-similarity was used to capture local terrain details.
• Gaussian process Bayesian optimization was employed for global accuracy.
• Perlin noise was integrated to improve terrain naturalness and detail.

Main Results:

• The proposed method significantly outperforms traditional interpolation techniques under sparse data conditions.
• Average errors were reduced by 30-40%.
• An R-squared value of 0.9836 was achieved, indicating high model accuracy.

Conclusions:

• The fractal-Gaussian process hybrid model offers a robust solution for high-resolution seafloor 3D reconstruction.
• This approach effectively handles complex seafloor terrain and sparse data limitations.
• The method enhances the capability for detailed seafloor topography and deformation monitoring.