Two-dimensional polarization distance underwater imaging method inspired by biomimetic vision

Area of Science:

• Optics
• Image Processing
• Marine Biology

Background:

• Polarization imaging offers advantages for underwater applications.
• Organisms use polarized vision for survival and navigation.
• Current algorithms struggle with underwater polarization data and scattered light estimation.

Purpose of the Study:

• To propose a 2D polarization distance model for underwater imaging.
• To quantify polarization differences between targets and background.
• To reduce estimation errors in background scattered light.

Main Methods:

• Developed a 2D polarization distance model combining Degree of Polarization (DOP) and Angle of Polarization (AOP).
• Utilized frequency domain analysis and polarization state reconstruction to determine background scattered light distribution.
• Quantified polarization differences as a mathematical distance.

Main Results:

• The proposed model effectively improves contrast between targets and background in turbid water.
• Reduced estimation errors of background scattered light.
• Demonstrated good visual enhancement effects.

Conclusions:

• The 2D polarization distance model is effective for underwater polarization imaging.
• This method enhances visibility and detail in challenging aquatic environments.
• The approach shows promise for improving underwater target detection and analysis.