Artificial Intelligence in Meta-optics

AI-Based Metamaterial Design

The use of metamaterials in various devices has revolutionized applications in optics, healthcare, acoustics, and power systems. Advancements in these fields demand novel or superior metamaterials that can demonstrate targeted control of electromagnetic, mechanical, and thermal properties of matter. Traditional design systems and methods often require manual manipulations which is time-consuming and resource intensive. The integration of artificial intelligence (AI) in optimizing metamaterial design can be employed to explore variant disciplines and address bottlenecks in design. AI-based metamaterial design can also enable the development of novel metamaterials by optimizing design parameters that cannot be achieved using traditional methods. The application of AI can be leveraged to accelerate the analysis of vast data sets as well as to better utilize limited data sets via generative models. This review covers the transformative impact of AI and AI-based metamaterial design for optics, acoustics, healthcare, and power systems. The current challenges, emerging fields, future directions, and bottlenecks within each domain are discussed.

Non-Invasive Self-Adaptive Information States' Acquisition inside Dynamic Scattering Spaces

Pushing the information states' acquisition efficiency has been a long-held goal to reach the measurement precision limit inside scattering spaces. Recent studies have indicated that maximal information states can be attained through engineered modes; however, partial intrusion is generally required. While non-invasive designs have been substantially explored across diverse physical scenarios, the non-invasive acquisition of information states inside dynamic scattering spaces remains challenging due to the intractable non-unique mapping problem, particularly in the context of multi-target scenarios. Here, we establish the feasibility of non-invasive information states' acquisition experimentally for the first time by introducing a tandem-generated adversarial network framework inside dynamic scattering spaces. To illustrate the framework's efficacy, we demonstrate that efficient information states' acquisition for multi-target scenarios can achieve the Fisher information limit solely through the utilization of the external scattering matrix of the system. Our work provides insightful perspectives for precise measurements inside dynamic complex systems.

Freeform metasurface color router for deep submicron pixel image sensors

Advances in imaging technologies have led to a high demand for ultracompact, high-resolution image sensors. However, color filter-based image sensors, now miniaturized to deep submicron pixel sizes, face challenges such as low signal-to-noise ratio due to fewer photons per pixel and inherent efficiency limitations from color filter arrays. Here, we demonstrate a freeform metasurface color router that achieves ultracompact pixel sizes while overcoming the efficiency limitations of conventional architectures by splitting and focusing visible light instead of filtering. This development is enabled by a fully differentiable topology optimization framework to maximize the use of the design space while ensuring fabrication feasibility and robustness to fabrication errors. The metasurface can distribute an average of 85% of incident visible light according to the Bayer pattern with a pixel size of 0.6 μm. The device and design methodology enable the compact, high-sensitivity, and high-resolution image sensors for various modern technologies and pave the way for the advanced photonic device design.

Research on underwater robot ranging technology based on semantic segmentation and binocular vision

Based on the principle of light refraction and binocular ranging, the underwater imaging model is obtained. It provides a theoretical basis for underwater camera calibration. In order to meet the requirement of underwater vehicle to identify and distance underwater target, a new underwater vehicle distance measurement system based on semantic segmentation and binocular vision is proposed. The system uses Deeplabv3 + to identify the underwater target captured by the binocular camera and generate the target map, which is then used for binocular ranging. Compared with the binocular ranging using the original drawing, the measurement accuracy of the proposed method has not changed, the measurement speed is increased by 30%, and the error rate is controlled within 5%, which meets the needs of underwater robot operations.

In Vivo Intelligent Fluorescence Endo-Microscopy by Varifocal Meta-Device and Deep Learning

Endo-microscopy is crucial for real-time 3D visualization of internal tissues and subcellular structures. Conventional methods rely on axial movement of optical components for precise focus adjustment, limiting miniaturization and complicating procedures. Meta-device, composed of artificial nanostructures, is an emerging optical flat device that can freely manipulate the phase and amplitude of light. Here, an intelligent fluorescence endo-microscope is developed based on varifocal meta-lens and deep learning (DL). The breakthrough enables in vivo 3D imaging of mouse brains, where varifocal meta-lens focal length adjusts through relative rotation angle. The system offers key advantages such as invariant magnification, a large field-of-view, and optical sectioning at a maximum focal length tuning range of ≈2 mm with 3 µm lateral resolution. Using a DL network, image acquisition time and system complexity are significantly reduced, and in vivo high-resolution brain images of detailed vessels and surrounding perivascular space are clearly observed within 0.1 s (≈50 times faster). The approach will benefit various surgical procedures, such as gastrointestinal biopsies, neural imaging, brain surgery, etc.

Nonlocal metasurface for dark-field edge emission

Nonlocal effects originating from interactions between neighboring meta-atoms introduce additional degrees of freedom for peculiar characteristics of metadevices, such as enhancement, selectivity, and spatial modulation. However, they are generally difficult to manipulate because of the collective responses of multiple meta-atoms. Here, we experimentally demonstrate the nonlocal metasurface to realize the spatial modulation of dark-field emission. Plasmonic asymmetric split rings (ASRs) are designed to simultaneously excite local dipole resonance and nonlocal quasi-bound states in the continuum and spatially extended modes. With one type of unit, nonlocal effects are tailored by varying array periods. ASRs at the metasurface's edge lack sufficient interactions, resulting in stronger dark-field scattering and thus edge emission properties of the metasurface. Pixel-level spatial control is demonstrated by simply erasing some units, providing more flexibility than conventional local metasurfaces. This work paves the way for manipulating nonlocal effects and facilitates applications in optical trapping and sorting at the nanoscale.

Machine Learning Aided Design and Optimization of Thermal Metamaterials

Artificial Intelligence (AI) has advanced material research that were previously intractable, for example, the machine learning (ML) has been able to predict some unprecedented thermal properties. In this review, we first elucidate the methodologies underpinning discriminative and generative models, as well as the paradigm of optimization approaches. Then, we present a series of case studies showcasing the application of machine learning in thermal metamaterial design. Finally, we give a brief discussion on the challenges and opportunities in this fast developing field. In particular, this review provides: (1) Optimization of thermal metamaterials using optimization algorithms to achieve specific target properties. (2) Integration of discriminative models with optimization algorithms to enhance computational efficiency. (3) Generative models for the structural design and optimization of thermal metamaterials.

Meta-Lens Particle Image Velocimetry

Fluid flow behavior is visualized through particle image velocimetry (PIV) for understanding and studying experimental fluid dynamics. However, traditional PIV methods require multiple cameras and conventional lens systems for image acquisition to resolve multi-dimensional velocity fields. In turn, it introduces complexity to the entire system. Meta-lenses are advanced flat optical devices composed of artificial nanoantenna arrays. It can manipulate the wavefront of light with the advantages of ultrathin, compact, and no spherical aberration. Meta-lenses offer novel functionalities and promise to replace traditional optical imaging systems. Here, a binocular meta-lens PIV technique is proposed, where a pair of GaN meta-lenses are fabricated on one substrate and integrated with a imaging sensor to form a compact binocular PIV system. The meta-lens weigh only 116 mg, much lighter than commercial lenses. The 3D velocity field can be obtained by the binocular disparity and particle image displacement information of fluid flow. The measurement error of vortex-ring diameter is ≈1.25% experimentally validates via a Reynolds-number (Re) 2000 vortex-ring. This work demonstrates a new development trend for the PIV technique for rejuvenating traditional flow diagnostic tools toward a more compact, easy-to-deploy technique. It enables further miniaturization and low-power systems for portable, field-use, and space-constrained PIV applications.

Matte surfaces with broadband transparency enabled by highly asymmetric diffusion of white light

The long-standing paradox between matte appearance and transparency has deprived traditional matte materials of optical transparency. Here, we present a solution to this centuries-old optical conundrum by harnessing the potential of disordered optical metasurfaces. Through the construction of a random array of meta-atoms tailored in asymmetric backgrounds, we have created transparent matte surfaces that maintain clear transparency regardless of the strength of disordered light scattering or their matte appearances. This remarkable property originates in the achievement of highly asymmetric light diffusion, exhibiting substantial diffusion in reflection and negligible diffusion in transmission across the entire visible spectrum. By fabricating macroscopic samples of such metasurfaces through industrial lithography, we have experimentally demonstrated transparent windows camouflaged as traditional matte materials, as well as transparent displays with high clarity, full color, and one-way visibility. Our work introduces an unprecedented frontier of transparent matte materials in optics, offering unprecedented opportunities and applications.

Broadband and large-aperture metasurface edge encoders for incoherent infrared radiation

The prevalence of computer vision systems necessitates hardware-based approaches to relieve the high computational demand of deep neural networks in resource-limited applications. One solution would be to off-load low-level image feature extraction, such as edge detection, from the digital network to the analog imaging system. To that end, this work demonstrates incoherent, broadband, low-noise optical edge detection of real-world scenes by combining the wavefront shaping of a 24-mm aperture metasurface with a refractive lens. An inverse design approach is used to optimize the metasurface for Laplacian-based edge detection across the 7.5- to 13.5-μm LWIR imaging band, allowing for facile integration with uncooled microbolometer-based LWIR imagers to encode edge information. A polarization multiplexed approach leveraging a birefringent metasurface is also demonstrated as a single-aperture implementation. This work could be applied to improve computer vision capabilities of resource-constrained systems by leveraging optical preprocessing to alleviate the computational requirements for high-accuracy image segmentation and classification.

Intelligent Phase Contrast Meta-Microscope System

Phase contrast imaging techniques enable the visualization of disparities in the refractive index among various materials. However, these techniques usually come with a cost: the need for bulky, inflexible, and complicated configurations. Here, we propose and experimentally demonstrate an ultracompact meta-microscope, a novel imaging platform designed to accomplish both optical and digital phase contrast imaging. The optical phase contrast imaging system is composed of a pair of metalenses and an intermediate spiral phase metasurface located at the Fourier plane. The performance of the system in generating edge-enhanced images is validated by imaging a variety of human cells, including lung cell lines BEAS-2B, CLY1, and H1299 and other types. Additionally, we integrate the ResNet deep learning model into the meta-microscope to transform bright-field images into edge-enhanced images with high contrast accuracy. This technology promises to aid in the development of innovative miniature optical systems for biomedical and clinical applications.

3D-printed multilayer structures for high-numerical aperture achromatic metalenses

Flat optics consisting of nanostructures of high-refractive index materials produce lenses with thin form factors that tend to operate only at specific wavelengths. Recent attempts to achieve achromatic lenses uncover a trade-off between the numerical aperture (NA) and bandwidth, which limits performance. Here, we propose a new approach to design high-NA, broadband, and polarization-insensitive multilayer achromatic metalenses (MAMs). We combine topology optimization and full-wave simulations to inversely design MAMs and fabricate the structures in low-refractive index materials by two-photon polymerization lithography. MAMs measuring 20 μm in diameter operating in the visible range of 400 to 800 nm with 0.5 and 0.7 NA were achieved with efficiencies of up to 42%. We demonstrate broadband imaging performance of the fabricated MAM under white light and RGB narrowband illuminations. These results highlight the potential of the 3D-printed multilayer structures for realizing broadband and multifunctional meta-devices with inverse design.

Flexible Thin Film Functionalized by Initiative Dust Removal and Anti-Fogging for Optical Device Applications

The deposition of dust and condensation of fog will block the scattering and transmission of light, thus affecting the performance of optical devices. In this work, flexible polyethylene terephthalate (PET) foil functionalized by active dust removal and anti-fogging characteristics is realized which combines electrodynamic screen (EDS) and electro-heating devices. In lieu of traditional measurement methods of dust removal efficiency, the PSNR is employed to characterize the dust removal efficiency of the film for the first time. The results show that both dust removal and anti-fogging improve the image quality, in which the dust removal increases the PSNR from 28.1 dB to 34.2 dB and the anti-fogging function realizes a film temperature rise of 16.7 ∘C in 5 min, reaching a maximum of 41.3 ∘C. According to the high sensitivity of the PSNR, we propose a fully automatic CIS film-driven algorithm, and its feasibility has been demonstrated.

Varifocal Metalenses: Harnessing Polarization-Dependent Superposition for Continuous Focal Length Control

Traditional varifocal lenses are bulky and mechanically complex. Emerging active metalenses promise compactness and design flexibility but face issues like mechanical tuning reliability and nonlinear focal length tuning due to additional medium requirements. In this work, we propose a varifocal metalens design based on superimposing light intensity distributions from two orthogonal polarization states. This approach enables continuous and precise focal length control within the visible spectrum, while maintaining relatively high focusing efficiencies (∼41% in simulation and ∼28% in measurement) and quality. In experimental validation, the metalens exhibited flexible tunability, with the focal length continuously adjustable between two spatial positions upon variation of the incident polarization angle. The MTF results showed high contrast reproduction and sharp imaging, with a Strehl ratio of >0.7 for all polarization angles. With compactness, design flexibility, and high focusing quality, the proposed varifocal metalens holds potential for diverse applications, advancing adaptive and versatile optical devices.

Deep learning enhanced achromatic imaging with a singlet flat lens

Correction of chromatic aberration is an important issue in color imaging and display. However, realizing broadband achromatic imaging by a singlet lens with high comprehensive performance still remains challenging, though many achromatic flat lenses have been reported recently. Here, we propose a deep-learning-enhanced singlet planar imaging system, implemented by a 3 mm-diameter achromatic flat lens, to achieve relatively high-quality achromatic imaging in the visible. By utilizing a multi-scale convolutional neural network (CNN) imposed to an achromatic multi-level diffractive lens (AMDL), the white light imaging qualities are significantly improved in both indoor and outdoor scenarios. Our experiments are fulfilled via a large paired imaging dataset with respect to a 3 mm-diameter AMDL, which guaranteed with achromatism in a broad wavelength range (400-1100 nm) but a relative low efficiency (∼45%). After our CNN enhancement, the imaging qualities are improved by ∼2 dB, showing competitive achromatic and high-quality imaging with a singlet lens for practical applications.

Meticulous research for design of plasmonics sensors for cancer detection and food contaminants analysis via machine learning and artificial intelligence

Cancer is one of the leading causes of death worldwide, making early detection and accurate diagnosis critical for effective treatment and improved patient outcomes. In recent years, machine learning (ML) has emerged as a powerful tool for cancer detection, enabling the development of innovative algorithms that can analyze vast amounts of data and provide accurate predictions. This review paper aims to provide a comprehensive overview of the various ML algorithms and techniques employed for cancer detection, highlighting recent advancements, challenges, and future directions in this field. The main challenge is finding a safe, auditable and reliable analysis method for fundamental scientific publication. Food contaminant analysis is a process of testing food products to identify and quantify the presence of harmful substances or contaminants. These substances can include bacteria, viruses, toxins, pesticides, heavy metals, allergens, and other chemical residues. Machine learning (ML) and artificial intelligence (A.I) proposed as a promising method that possesses excellent potential to extract information with high validity that may be overlooked with conventional analysis techniques and for its capability in a wide range of investigations. A.I technology used in meta-optics can develop optical devices and systems to a higher level in future. Furthermore (M.L.) and (A.I.) play key roles as a health Approach for nano materials NMs safety assessment in environment and human health research. Beside, benefits of ML in design of plasmonic sensors for different applications with improved resolution and detection are convinced.

Miniature Two-Photon Microscopic Imaging Using Dielectric Metalens

Miniature two-photon microscopy has emerged as a powerful technique for investigating brain activity in freely moving animals. Ongoing research objectives include reducing probe weight and minimizing animal behavior constraints caused by probe attachment. Employing dielectric metalenses, which enable the use of sizable optical components in flat device structures while maintaining imaging resolution, is a promising solution for addressing these challenges. In this study, we designed and fabricated a titanium dioxide metalens with a wavelength of 920 nm and a high aspect ratio. Furthermore, a meta-optic two-photon microscope weighing 1.36 g was developed. This meta-optic probe has a lateral resolution of 0.92 μm and an axial resolution of 18.08 μm. Experimentally, two-photon imaging of mouse brain structures in vivo was also demonstrated. The flat dielectric metalens technique holds promising opportunities for high-performance integrated miniature nonlinear microscopy and endomicroscopy platforms in the biomedical field.

Research Progress in Surface-Enhanced Infrared Absorption Spectroscopy: From Performance Optimization, Sensing Applications, to System Integration

Infrared absorption spectroscopy is an effective tool for the detection and identification of molecules. However, its application is limited by the low infrared absorption cross-section of the molecule, resulting in low sensitivity and a poor signal-to-noise ratio. Surface-Enhanced Infrared Absorption (SEIRA) spectroscopy is a breakthrough technique that exploits the field-enhancing properties of periodic nanostructures to amplify the vibrational signals of trace molecules. The fascinating properties of SEIRA technology have aroused great interest, driving diverse sensing applications. In this review, we first discuss three ways for SEIRA performance optimization, including material selection, sensitivity enhancement, and bandwidth improvement. Subsequently, we discuss the potential applications of SEIRA technology in fields such as biomedicine and environmental monitoring. In recent years, we have ushered in a new era characterized by the Internet of Things, sensor networks, and wearable devices. These new demands spurred the pursuit of miniaturized and consolidated infrared spectroscopy systems and chips. In addition, the rise of machine learning has injected new vitality into SEIRA, bringing smart device design and data analysis to the foreground. The final section of this review explores the anticipated trajectory that SEIRA technology might take, highlighting future trends and possibilities.

Bright-Field and Edge-Enhanced Imaging Using an Electrically Tunable Dual-Mode Metalens

The imaging of microscopic biological samples faces numerous difficulties due to their small feature sizes and low-amplitude contrast. Metalenses have shown great promise in bioimaging as they have access to the complete complex information, which, alongside their extremely small and compact footprint and potential to integrate multiple functionalities into a single device, allow for miniaturized microscopy with exceptional features. Here, we design and experimentally realize a dual-mode metalens integrated with a liquid crystal cell that can be electrically switched between bright-field and edge-enhanced imaging on the millisecond scale. We combine the concepts of geometric and propagation phase to design the dual-mode metalens and physically encode the required phase profiles using hydrogenated amorphous silicon for operation at visible wavelengths. The two distinct metalens phase profiles include (1) a conventional hyperbolic metalens for bright-field imaging and (2) a spiral metalens with a topological charge of +1 for edge-enhanced imaging. We demonstrate the focusing and vortex generation ability of the metalens under different states of circular polarization and prove its use for biological imaging. This work proves a method for in vivo observation and monitoring of the cell response and drug screening within a compact form factor.

Metasurface designed with quantitative field distributions

A new method for designing metasurfaces has been proposed and demonstrated, which allows for the generation of precise quantitative field distributions. This unique approach involves combining a tandem neural network with an iterative algorithm to optimize the metasurface design, enabling accurate control over the intensity and polarization of the resulting field. This strategy is both efficient and robust and has the potential to accelerate the development of metasurface devices with complex functionalities.

Broadband Solar Metamaterial Absorbers Empowered by Transformer-Based Deep Learning

The research of metamaterial shows great potential in the field of solar energy harvesting. In the past decade, the design of broadband solar metamaterial absorber (SMA) has attracted a surge of interest. The conventional design typically requires brute-force optimizations with a huge sampling space of structure parameters. Very recently, deep learning (DL) has provided a promising way in metamaterial design, but its application on SMA development is barely reported due to the complicated features of broadband spectrum. Here, this work develops the DL model based on metamaterial spectrum transformer (MST) for the powerful design of high-performance SMAs. The MST divides the optical spectrum of metamaterial into N patches, which overcomes the severe problem of overfitting in traditional DL and boosts the learning capability significantly. A flexible design tool based on free customer definition is developed to facilitate the real-time on-demand design of metamaterials with various optical functions. The scheme is applied to the design and fabrication of SMAs with graded-refractive-index nanostructures. They demonstrate the high average absorptance of 94% in a broad solar spectrum and exhibit exceptional advantages over many state-of-the-art counterparts. The outdoor testing implies the high-efficiency energy collection of about 1061 kW h m^-2 from solar radiation annually. This work paves a way for the rapid smart design of SMA, and will also provide a real-time developing tool for many other metamaterials and metadevices.