Infrared Target Reconstruction Under Detector Multiplexing Using Polarization Encoding and Stokes Vector Decoding

Area of Science:

• Optics and Photonics
• Infrared Imaging Technology
• Polarization Optics

Background:

• Wide-field infrared systems face limitations in detector size and cooling.
• Current multiplexing techniques suffer from spatial aliasing, degrading performance.
• Need for efficient spatial information recovery in compact sensing applications.

Purpose of the Study:

• To propose a polarization-encoded field-of-view (FOV) multiplexing method.
• To recover spatial information from aliased detector measurements in infrared imaging.
• To enhance target localization performance in multi-FOV systems.

Main Methods:

• Dividing the imaging plane into distinct polarization-encoded FOV regions.
• Superimposing modulated sub-images onto a common detector via optical folding.
• Reconstructing pixel-wise Stokes vectors and identifying spatial origins using polarization-domain matching and voting.

Main Results:

• Achieved 97.3% pixel-level classification accuracy under ideal conditions.
• Maintained over 95% accuracy with noise levels of σ = 0.02.
• Demonstrated low normalized Stokes reconstruction error (<0.02) and stable performance within ±10° polarization modulation deviations.

Conclusions:

• Polarization encoding enables efficient separation of multiplexed spatial information.
• The method avoids increasing detector resources for compact wide-field infrared sensing.
• Potential for improved performance in constrained infrared imaging applications.