Comparing Machine Learning and Physics-Based Nanoparticle Geometry Determinations Using Far-Field Spectral Properties

Abstract

Anisotropic metal nanostructures exhibit polarization-dependent light scattering. This property has been widely exploited to determine geometries of subwavelength structures using far-field microscopy. Here, we explore the use of variational autoencoders (VAEs) to determine the geometries of gold nanorods (NRs) such as in-plane orientation and aspect ratio under linearly polarized dark-field illumination in an optical microscope. We input polarized dark-field scattering spectra and electron microscopy images into a dual-branch multimodal VAE with a single shared latent space trained on paired spectra-image data, using a learnable linear adapter. We achieve prediction of Au NRs using only polarized dark-field scattering spectra input. We determine geometrical parameters of orientational angle and aspect ratio quantitatively via both dual-VAE and physics-based analysis. We show that orientational angle prediction by dual-VAE performs well with only a small (300 particle) training set, yielding a mean absolute error (MAE) of 14.4 and a concordance correlation coefficient (CCC) of 0.95. This performance is only marginally worse than the physics-based cos(2theta) fitting approach between the scattering intensity and the polarizing angle, which achieves MAE of 8.78 and CCC of 0.99. Aspect ratio determination is also similar for the dual-VAE and physics-based fitting comparison (MAE of 0.21 vs. 0.23 and CCC of 0.53 vs. 0.68). By learning a shared latent manifold linking spectra and morphology, the model can generate NR images with accurate orientation and aspect ratio with spectra-only input in the small-data regime (300 particles), suggesting a general recipe for inverse nano-optical problems requiring both structure and orientation information.