Underwater single-polarization descattering for targets with complex polarization properties

Area of Science:

• Optics and Photonics
• Remote Sensing
• Image Processing

Background:

• Underwater polarization imaging recovers target information in scattering media.
• Traditional methods require multiple polarization images, leading to complex systems.

Purpose of the Study:

• To develop a simplified underwater polarization imaging method.
• To improve image quality in turbid water using a single-input polarization technique.

Main Methods:

• Solving for the perpendicular-polarized image from the parallel-polarized image using generalized Malus' law.
• Employing an adaptive weighting mechanism for non-uniform target light intensity.
• Using an adaptive Gaussian low-pass filter for backscattered light estimation and separation.

Main Results:

• Significantly improved image quality for complex polarization targets in turbid water.
• Outperformed existing single-polarization descattering techniques and traditional methods.
• Demonstrated effective separation of target light from backscattered light.

Conclusions:

• The proposed method enables compact underwater polarization imaging systems.
• Provides theoretical insights and technical support for advanced underwater imaging.
• Advances the field of polarization imaging in scattering environments.