Polarization-informed deep learning for 3D integral imaging restoration in turbidity

Research context

Research on electrical energy conversion, storage and generation dates back to the nineteenth century, but only in recent years have scientists begun to investigate the impact of electron spin on these processes. The ability to control and manipulate this intrinsically quantum property of matter opens new approaches to addressing energy science challenges. The chiral-induced spin selectivity (CISS) effect is central to this effort, as it enables control over the transport and generation of both pure spin currents and spin-polarized charge currents. In this Review, we first introduce design strategies for implementing CISS in materials and then describe examples of how CISS has been used to improve electrocatalysis and spintronics. We conclude with a forward-looking perspective on the next steps for leveraging CISS in energy science.

Keywords: Mueller calculus, Polarimetry, Stokes parameters, Integral imaging, Convolutional neural network, Deep learning, Matrix (chemical analysis), Polarization (electrochemistry), Artificial intelligence, Computer science, Optics, Computer vision, Remote sensing, Radar imaging, Ghost imaging, Image restoration, Pattern recognition (psychology), Iterative reconstruction, Artificial neural network, Synthetic aperture radar, Transmission (telecommunications), Degree of polarization, Image processing, Scattering

Source & review

Bibliographic record reviewed for relevance and publication quality. Full-text findings have not been extracted; consult the original publication for methods and results.

OpenAlex cited-by 6; topical title/abstract and venue audit passed.

6 citations · OpenAlex · observed 2026-09-08

Metadata: OpenAlex · source record ↗

Cite / 引用

Gokul Krishnan, Bishwanath Pal, Jiheon Lee, Kashif Usmani, Saurabh Goswami, Bahram Javidi. Polarization-informed deep learning for 3D integral imaging restoration in turbidity. Optics Express (2025). https://doi.org/10.1364/oe.579903

分享：本论文的永久链接