Disordered metasurface enabled single-shot full-Stokes polarization imaging leveraging weak dichroism

Abstract

Polarization, one of the fundamental properties of light, is critical for certain imaging applications because it captures information from the scene that cannot directly be recorded by traditional intensity cameras. Currently, mainstream approaches for polarization imaging rely on strong dichroism of birefringent crystals or artificially fabricated structures that exhibit a high diattenuation typically exceeding 99%, which corresponds to a polarization extinction ratio (PER) >~100. This not only limits the transmission efficiency of light, but also makes them either offer narrow operational bandwidth or be non-responsive to the circular polarization. Here, we demonstrate a single-shot full-Stokes polarization camera incorporating a disordered metasurface array with weak dichroism. The diattenuation of the metasurface array is ~65%, which corresponds to a PER of ~2. Within the framework of compressed sensing, the proposed disordered metasurface array serves as an efficient sensing matrix. By incorporating a mask-aware reconstruction algorithm, the signal can be accurately recovered with a high probability. In our experiments, the proposed approach exhibits high-accuracy full-Stokes polarimetry and high-resolution real-time polarization imaging. Our demonstration highlights the potential of combining meta-optics with reconstruction algorithms as a promising approach for advanced imaging applications.

Fig. 1. Disordered dielectric metasurface array with weak dichroism.

a Metasurface element composed of a two-dimensional array of polarization-dependent meta-pixels. b The operational principle of a single meta-pixel. Each meta-pixel mapping to the sensor pixels is sensitive to two arbitrary orthogonal SoPs [${\stackrel{\rightharpoonup }{\alpha }}^{+}$,${\stackrel{\rightharpoonup }{\alpha }}^{-}$], including linear and circular polarizations. I represents the light intensity. ${\mu }_1$ and ${\mu }_2$ denote the transmission coefficients. c–e Experimentally measured transmission of three representative meta-pixels with operation wavelengths covering the entire visible range from 400 nm to 700 nm. f Optical microscope image of the fabricated metasurface device. The inset shows a zoomed-in meta-pixel array. g, h The SEM images show the top view and oblique view of the TiO₂ nanopillars.

Fig. 2. The operational principle of the nanophotonic polarization camera consists of three steps: calibration, capture, and reconstruction.

a Schematic of the calibration process. ➀ denotes a narrowband spectral filter. ➁ denotes an imaging lens. b The responsivity matrix $\mathit{R}$ of the whole meta-device is calibrated from the photoresponses to the three pairs of orthogonal SoPs of incident light. c Schematic of the capture process. Scattered optical waves at each position (x, y) of target scene represented an unknown SoP $\mathit{S}={\left[S_0,S_1,S_2,S_3\right]}^T$, illuminates the imaging system. d A single-shot capture process generates the polarization-encoded image I. e Mathematical description of the relation between ${\mathit{S}}_{\mathit{v}}$, I_v, and $\mathit{\Lambda }$. f SoP reconstruction by the convolutional neural network. g Reconstructed SoP and corresponding position on Poincaré sphere.

Fig. 3. Comparison of sampling efficiency between ordered and disordered designs.

a–c Ordered polarization filter arrangements and the corresponding volume enclosed within the Poincaré sphere. The red box denotes the basic arrangement unit. The coordinate axes represent the last three components of the Stokes-like vector ${\mathit{M}}_{i,j}={\left[m_{i,j}^0,m_{i,j}^1,m_{i,j}^2,m_{i,j}^3\right]}^T$. d $\mu \left(\mathit{\Lambda },\mathit{\Psi }\right)$ on different arrangements. Due to variations in sampling efficiency within each 12 × 12 patch of the designed disordered metasurface array, we randomly select 10,000 patches from our 400 × 400 metasurface array to calculate the sampling efficiency and illustrate these results in the form of density plot for fairness. The inset in (d) is a 12 × 12 patch, serving as an example to illustrate the arrangement of random polarization encoding.

Fig. 4. Full-Stokes polarimetric measurements.

a Three representative SoPs are chosen for polarization measurement. b Comparison of the SoPs obtained using a commercial polarimeter (green solid lines) and our method (red dots), using polar plots and polarization ellipses, at an operational wavelength of 550 nm. The radius on the polar plot indicates the normalized light intensity. Blue arrows denote the handedness of light. c The reconstruction errors of Stokes parameters (S₁, S₂, S₃) from 25 arbitrarily selected SoPs on Poincaré sphere.

Fig. 5. Full-Stokes polarization imaging.

Indoor (a–c) and outdoor (d) photography with the proposed metasurface camera. In each case, the raw exposure (corresponding to the polarization-encoded image, or called compressed image), S₀ (corresponding to the monochrome intensity image), the azimuth of the polarization ellipse and the degree of polarization are shown.

Fig. 6. Polarization imaging of S₃.

In each example, raw sensor acquisition, S₀, S₃ and DoP are shown. a Biological specimen. b 3D glasses consist of opposite circular polarization filters. c A rectangle acrylic piece is not stressed by the clamp-squeezing and displays no stress-induced birefringence in the S₃ image. d The rectangle acrylic piece is stressed by clamp-squeezing and displays stress-induced birefringence in the S₃ image.