Polarization-adaptive scattering suppression method for underwater structured-light 3D reconstruction

Area of Science:

• Robotics and Computer Vision
• Optical Engineering
• Fluid Dynamics

Background:

• Underwater 3D reconstruction is challenged by particle scattering, degrading image quality and accuracy.
• Existing methods often fail to preserve essential fringe data for structured-light systems.
• Current techniques lack adaptability to varying scattering conditions and require manual input.

Purpose of the Study:

• To develop an adaptive underwater descattering method preserving fringe characteristics for structured-light 3D reconstruction.
• To enhance the accuracy and robustness of 3D reconstruction in scattering environments.
• To overcome limitations of existing 2D enhancement and manual-dependent methods.

Main Methods:

• Proposed a polarization-phase separation and adaptive descattering (PSAD) method.
• Utilized temporal orthogonal decoupling to preserve scattering field properties.
• Employed low-rank sparse decomposition for adaptive polarization estimation.
• Implemented adaptive iteration for background illumination and transmittance estimation.

Main Results:

• PSAD effectively suppresses underwater scattering while preserving fringe modulation.
• Achieved significant optimization in point cloud quality.
• Reduced the root mean square error (RMSE) of 3D reconstruction by up to 69.7%.
• Demonstrated stable, high-precision 3D reconstruction in diverse turbidity levels.

Conclusions:

• The PSAD method offers a robust solution for high-precision underwater 3D reconstruction.
• It significantly improves reconstruction accuracy compared to existing approaches.
• Enables reliable 3D mapping in challenging, scattering underwater conditions.