Image reconstruction of underwater objects using regression-estimated remote Mueller matrix polarimetry
Optics Express
|August 14, 2026
Summary
This study introduces a polarimetric imaging method to reconstruct clear, air-equivalent images of underwater objects. It uses Mueller matrix analysis and regression to reduce water-induced polarization distortions for better remote sensing.
Area of Science:
- Optics and Photonics
- Image Processing
- Remote Sensing
Background:
- The Mueller matrix completely describes light polarization changes in a medium.
- Underwater imaging is challenged by medium-induced polarization distortions.
- Existing methods struggle with accurate polarization compensation in scattering environments.
Purpose of the Study:
- To develop a polarimetric imaging approach for reconstructing air-equivalent images of submerged objects.
- To reduce polarization distortions caused by the underwater medium.
- To enable clearer remote sensing and operation in scattering underwater environments.
Main Methods:
- Combining Mueller matrix analysis with regression modeling.
- Learning the relationship between polarization measurements in air and through water.
- Estimating the Mueller matrix of the intervening water medium without direct access.
Main Results:
- Successful reconstruction of air-equivalent images of submerged objects.
- Significant reduction in medium-induced polarization distortions.
- Demonstrated capability for remote object reconstruction in challenging conditions.
Conclusions:
- The developed polarimetric imaging approach effectively compensates for underwater polarization distortions.
- This method offers a practical strategy for enhanced underwater imaging, sensing, and operation.
- Enables the remote reconstruction of previously unseen underwater objects.
