Mueller matrix based imaging enabled by polarization demultiplexing metasurfaces

This study presents a polarization-based approach to reconstruct the air-equivalent image of an object submerged in water using Mueller matrix analysis. By isolating and removing the medium’s optical effects, we demonstrate accurate recovery of the object’s appearance even though imaged under varying turbidity conditions. The results show that only specific Mueller matrix components are sensitive to changes in water turbidity, enabling the potential for remote prediction and correction. A metasurface with polarization demultiplexing capabilities can further simplify the system by eliminating the need for rotating filters, offering a compact, time-efficient solution for real-time underwater imaging and sensing applications.