Label-free deeply subwavelength optical microscopy

Abstract

We report the experimental demonstration of deeply subwavelength far-field optical microscopy of unlabelled samples. We beat the ~λ/2 diffraction limit of conventional optical microscopy several times over by recording the intensity pattern of coherent light scattered from the object into the far-field. We retrieve information about the object with a deep learning neural network trained on scattering events from a large set of known objects. The microscopy retrieves dimensions of the imaged object probabilistically. Widths of the sub-wavelength components of the dimer are measured with precision of λ/10 with probability higher that 95% and with precision of λ/20 with probability of better that 77%. The ability to image at the nanometer scale using visible light remains a long-standing fundamental challenge for optics. Despite over 400 years of developments in microscopy, subwavelength optical imaging is only possible through the use of near-field probes [1] or fluorescent labels [2,3]. Moreover, combining the latter with artificial intelligence approaches has shown to improve imaging resolution [4,5]. However, fluorescence-based and near-field methods exlude many applications. Several other techniques have been developed to break through the "diffraction limit" of conventional microscopes [6-8], which however led only to modest enhancement of resolution in far-field techniques [9,10]. The distinction between far-field and near-field imaging techniques is important. In the context of imaging, the near-field and far-field zones are regions of the electromagnetic field around an object, resulting from radiation scattering on the object. The near-field consists of non-A

Experimental realization of the active convolved illumination imaging technique for enhanced signal-to-noise ratio

“…Optimal and precise manipulation of the pupil function, beyond conventional optics, may be possible with the advent of amenable metasurfaces [16][17][18]20]. More work on ACI and potential relevance with structured illumination microscopy [3,[49][50][51][52][53], superoscillatory imaging [79][80][81][82][83], and super-gain [84,85] can enrich further understanding of imaging beyond the known boundaries. Computational superresolution techniques [42][43][44][58][59][60] integrated with ACI can provide extended resolution limits.…”

Artificial intelligence for photonics and photonic materials

“…Furthermore, Pu et al [65,81] (Figure 4c) have shown that the combination of machine learning with photonics can revolutionize one of the most important field in optics, imaging. In a theoretical article [65], they first introduced a new imaging technique termed Deeply Subwavelength Superoscillatory Imaging (DSSI) that has the potential to reveal the fine structure of a physical object through its farfield scattering pattern under superoscillatory illumination with a resolution far beyond the conventional "diffraction limit" exceeding λ/200 for a dimer comprising two subwavelength opaque particles.…”

“…In a theoretical article [65], they first introduced a new imaging technique termed Deeply Subwavelength Superoscillatory Imaging (DSSI) that has the potential to reveal the fine structure of a physical object through its farfield scattering pattern under superoscillatory illumination with a resolution far beyond the conventional "diffraction limit" exceeding λ/200 for a dimer comprising two subwavelength opaque particles. Then, they have demonstrated this new imaging technique experimentally [81] by retrieving the parameters of a physical object from its scattering pattern with resolution exceeding λ/20. And in an application of AI to optical metrology, Rendón-Barraza et al [82] have demonstrated how the physical size of subwavelength objects can be determined with accuracy exceeding λ/800 via a deep learning-enabled analysis of diffraction patterns.…”

Unlabeled Far‐Field Deeply Subwavelength Topological Microscopy (DSTM)

“…Very recently, it was demonstrated that far‐field, noncontact, label‐free, optical, high‐resolution imaging can be achieved by analyzing intensity patterns of light scattered by the object using artificial intelligence. [ 5,6 ] Here we demonstrate by virtue of numerical and proof‐of‐principle real‐life experiments that further improvement in resolution can be achieved by illuminating the object with topologically structured light. The improved resolution results from the interactions of the object's fine features with singularities of highly structured topological light.…”