Polarization is a fundamental aspect of light waves that plays a crucial role in numerous optical applications. From sunglasses and camera lenses to advanced optical communication and imaging systems, understanding and mastering the polarization of light is vital for the advancement of these technologies. However, manipulating the spatial distribution of the polarization state of an optical field has proven to be challenging. Existing polarization modulation devices are limited in their range of polarization scattering matrices and are ineffective for unpredictable spatially varying polarization fields.
A research team at UCLA has introduced a groundbreaking design for a polarization transformer diffractive network. This new approach opens up possibilities for universal polarization transformations of spatially varying polarization fields. Detailed in the journal Advanced Materials, the diffractive network is capable of all-optically synthesizing complex-valued polarization scattering matrices between different locations within its input and output fields-of-views.
The diffractive polarization transformer is composed of a series of isotropic diffractive layers, each coded with subwavelength-level diffractive features. These layers have optimizable transmission coefficients. Linear polarizers are also utilized within the diffractive processor volume, orientated at different angles and remaining fixed. Through supervised deep learning, the diffractive polarization transformer is trained to implement a large set of arbitrarily-selected polarization scattering matrices.
The feasibility of this diffractive polarization transformer was confirmed through experiments conducted by the UCLA research team. Using wire-grid polarizers and 3D-printed diffractive layers, a proof-of-concept design was built to perform a user-defined polarization permutation operation in the terahertz part of the electromagnetic spectrum. The fabricated diffractive polarization transformer successfully implemented various spatially-encoded polarization scattering matrices, effectively navigating the input polarization states to the desired spatially varying polarization states at the output.
The UCLA team plans to further enhance the design of their diffractive polarization transformer for operation under broadband illumination. This would allow for the simultaneous processing of multiple features encoded in optical fields, including amplitude, phase, polarization, and spectral features. These advancements could lead to the development of intelligent machine vision systems with polarization-aware object detection and classification capabilities. Such systems could have applications in remote sensing, security/defense, material inspection, and medical imaging.
The introduction of the diffractive polarization transformer by the research team at UCLA offers a promising solution to the challenges in manipulating the spatial distribution of polarization in light waves. With its ability to implement a wide range of polarization scattering matrices and navigate unpredictable spatially varying polarization fields, this new approach has the potential to revolutionize various optical technologies. As further improvements are made, diffractive polarization transformers could pave the way for advanced machine vision systems and significant advancements in the fields of remote sensing, security, material inspection, and medical imaging.
Leave a Reply