Artificial neural networks have shown great proficiency in transforming low-resolution microscopic images into high-resolution images. However, training data remains a challenge, as large-scale open-source databases of microscopic images are rare, particularly 3D data. Moreover, the long training times and the need for expensive computational resources have become a burden to the research community. We introduced a deep-learning-based self-supervised volumetric imaging approach, which we termed “Self-Vision.” The self-supervised approach requires no training data, apart from the input image itself. The lightweight network takes just minutes to train and has demonstrated resolution-enhancing power on par with or better than that of a number of recent microscopy-based models. Moreover, the high throughput power of the network enables large image inference with less postprocessing, facilitating a large field-of-view (2.45mm×2.45mm) using a home-built two-photon microscopy system. Self-Vision can recover images from fourfold undersampled inputs in the lateral and axial dimensions, dramatically reducing the acquisition time. Self-Vision facilitates the use of a deep neural network for 3D microscopy imaging, easing the demanding process of image acquisition and network training for current resolution-enhancing networks.

High-order Bessel beams are of great interest for most stable long-range optical quantum communications due to their unique nondiffraction, self-healing, and orbital angular-momentum-carrying capabilities. Until now, metasurfaces based on Bessel beam generators are mostly static and focused on generating zero-order Bessel beams. A moiré meta-device made of two cascaded metasurfaces is a simple, effective strategy to dynamically manipulate the wavefront of electromagnetic (EM) waves by mutual rotation between the two metasurfaces. Here, an all-dielectric moiré meta-device integrated with the functions of an axicon and a spiral phase plate to generate terahertz Bessel beams is designed. Not only the order, but also the nondiffraction length of the generated Bessel beam can be continuously tuned. As a proof of concept of the feasibility of the platform, the case of tuning order is experimentally demonstrated. The experimental results are in good agreement with the theoretical expectations. In addition, we also numerically proved that the nondiffraction length of the Bessel beam can be adjusted with the same approach. The moiré meta-device platform is powerful in dynamically manipulating the wavefront of EM waves and provides an effective strategy for continuously controlling the properties of the Bessel beam, which may find applications in optical communications, particle manipulation, and super-resolution imaging.

Full-color micro-LED displays are being widely developed and regarded as a primary option in current microdisplay technologies to fulfill the urgent demands of metaverse applications in the next decade. In this paper, a monolithic full-color micro-LED microdisplay with a resolution of 423 pixels per inch is demonstrated through the integration of a blue GaN-on-Si display module and a quantum dots photoresist (QDs-PR) color conversion module. The 400×240 active-matrix blue micro-LED display with a dominant wavelength of 440 nm was monolithically fabricated using GaN-on-Si epiwafers and flip-chip bonded on a custom-designed complementary metal-oxide semiconductor backplane. A color conversion module was independently fabricated on a 4-in. sapphire substrate by applying red and green QDs-PR arrays and a color filter array through the standard lithography process. Combining the blue GaN-on-Si micro-LED display module and the lithography-based QDs-PR color conversion module, a full-color micro-LED display was achieved with a wide color gamut up to 104% of the standard red, green, and blue and a maximum brightness of over 500 nits. The influence of blue light leakage resulting from the possible misalignment of flip-chip bonding and crosstalk in the bottom GaN-on-Si display was investigated in which the percentages of efficient pumping light for the blue, green, and red subpixels are around 95%, 89%, and 92%, respectively. This prototype demonstrates potential scalability and low-cost volume production of high-resolution full-color micro-LED microdisplays soon.

Metallic nanoplasmonics, due to its extremely small size and ultrafast speed, has been one of the key components for next-generation information technology. It is vital that the highly tunable nanoplasmonic system in the solid state can be achieved for optoelectronic devices, which, still remains elusive for the visible region. Here we sandwich vanadyl oxalate (VOC2O4) thin films in-between gold nanoparticles and gold film to establish thermo-responsive nanoantennas. The thickness of the VOC2O4 composite films remains almost unchanged within the temperature cycles between 15°C and 80°C, while the refractive index of the films decreases with the increase of temperature due to the dehydration, which results in blueshift of the plasmon peak up to 60 nm. The plasmon resonances can be fully recovered when the temperature cools down again. This process is reversible within the temperature range of 15°C–80°C, which can be optically modulated with photothermal effect. Such thermo-responsive plasmonic nanoantenna works in the solid state with hundreds of kilohertz switching speed, which is highly compatible with traditional optoelectronic devices. It can be envisioned that this thermo-responsive optical thin film can be a promising candidate for integrated nanoplasmonic and optoelectronic devices.

Structured optical fields embedded with polarization singularities (PSs) have attracted extensive attention due to their capability to retain topological invariance during propagation. Many advances in PS research have been made over the past 20 years in the areas of mathematical description, generation and detection technologies, propagation dynamics, and applications. However, one of the most crucial and difficult tasks continues to be manipulating PSs with multiple degrees of freedom, especially in three-dimensional (3D) tailored optical fields. We propose and demonstrate the longitudinal PS lines obtained by superimposing Bessel-like modes with orthogonal polarization states on composite vector optical fields (VOFs). The embedded PSs in the fields can be manipulated to propagate robustly along arbitrary trajectories, or to annihilate, revive, and transform each other at on-demand positions in 3D space, allowing complex PS’ topological morphology and intensity patterns to be flexibly customized. Our findings could spur further research into singular optics and help with applications such as micromanipulation, microstructure fabrication, and optical encryption.

Optically transparent microwave absorbing metasurfaces have shown great potential and are needed in multiple applications environments containing optical windows owing to their ability to reduce backscattering electromagnetic (EM) signals while keeping continuous optical observation. Meanwhile, they are also required to have adaptive EM manipulation capability to cope with complex and capricious EM environments. As a general approach, distributed circuit components, including positive-intrinsic-negative diodes and varactors and sensing components, are integrated with passive absorbing metasurfaces to realize adaptive control of microwave absorption. However, these circuit elements generally require bulky electrical wires and complex control circuits to regulate the operating state, resulting in the absorbing structures being optically opaque. Hence, it is a great challenge to realize self-operating absorbers while maintaining optical transparency. Here, we report an optically transparent cognitive metasurface made of patterned graphene sandwich structures and a radio frequency detector, which can achieve adaptive frequency manipulation to match incident EM waves. As a proof-of-principle application example, we realize a closed-loop automatic absorber system prototype of the proposed graphene metasurface with self-adaptive frequency variation, without any human intervention. The approach may facilitate other adaptive metadevices in microwave regime with high-level recognition and manipulation and, more generally, promote the development of intelligent stealth technologies.

Raman fiber lasers (RFLs) have broadband tunability due to cascaded stimulated Raman scattering, providing extensive degrees of freedom for spectral manipulation. However, the spectral diversity of RFLs depends mainly on the wavelength flexibility of the pump, which limits the application of RFLs. Here, a spectrally programmable RFL is developed based on two-dimensional spatial-to-spectral mapping of light in multimode fibers (MMFs). Using an intracavity wavefront shaping method combined with genetic algorithm optimization, we launch light with a selected wavelength(s) at MMF output into the active part of the laser for amplification. In contrast, the light of undesired wavelengths is blocked. We demonstrate spectral shaping of the high-order RFL, including a continuously tunable single wavelength and multiple wavelengths with a designed spectral shape. Due to the simultaneous control of different wavelength regions, each order of Raman Stokes light allows flexible and independent spectral manipulation. Our research exploits light manipulation in a fiber platform with multi-eigenmodes and nonlinear gain, mapping spatial control to the spectral domain and extending linear light control in MMFs to active light emission, which is of great significance for applications of RFLs in optical imaging, sensing, and spectroscopy.

In this paper, we present an approach called the free lens modulation (FLM) method to generate high-perfection 3D generalized perfect optical vortices (GPOVs) with topological charges of 1–80. In addition, 2D and 3D GPOVs were produced by altering the parameters of the freely shaped lenses. To verify the quality of the GPOVs produced with the FLM method, we conducted optical trapping experiments and realized linear control of the rotation rate of the trapped particle. Due to the great advantages of high perfection and high power usage in generating arbitrarily shaped GPOVs, the FLM method is expected to be applied in optical manipulation, optical communications, and other fields.

Characterization of the state of polarization (SOP) of ultrafast laser emission is relevant in several application fields such as field manipulation, pulse shaping, testing of sample characteristics, and biomedical imaging. Nevertheless, since high-speed detection and wavelength-resolved measurements cannot be simultaneously achieved by commercial polarization analyzers, single-shot measurements of the wavelength-resolved SOP of ultrafast laser pulses have rarely been reported. Here, we propose a method for single-shot, wavelength-resolved SOP measurements that exploits the method of division-of-amplitude under far-field transformation. A large accumulated chromatic dispersion is utilized to time-stretch the laser pulses via dispersive Fourier transform, so that spectral information is mapped into a temporal waveform. By calibrating our test matrix with different wavelengths, wavelength-resolved SOP measurements are achieved, based on the division-of-amplitude approach, combined with high-speed opto-electronic processing. As a proof-of-concept demonstration, we reveal the complex wavelength-dependent SOP dynamics in the build-up of dissipative solitons. The experimental results show that the dissipative soliton exhibits far more complex wavelength-related polarization dynamics, which are not shown in single-shot spectrum measurement. Our method paves the way for single-shot measurement and intelligent control of ultrafast lasers with wavelength-resolved SOP structures, which could promote further investigations of polarization-related optical signal processing techniques, such as pulse shaping and hyperspectral polarization imaging.

Dynamic beam steering with unlimited angular range and fast speed remains a challenge in the terahertz gap, which is urgently needed for next-generation target tracking, wireless communications, and imaging applications. Different from metasurface phased arrays with element-level phase control, here we steer the beam by globally engineering the diffraction of two cascaded metagratings during in-plane rotation. Benefiting from large-angle diffraction and flexible on/off control of the diffraction channels, a pair of metagratings with optimized supercells and proper orientation successfully directs the incoming beam towards any arbitrary direction over the transmission half space, with the steering speed improved more than twice that of the small-angle diffractive designs. Single-beam and dual-beam steering within the solid angle of 1.56π and elevation angle of ±77° has been demonstrated with average throughput efficiency of 41.4% at 0.14 THz, which can be generalized to multiple-beam cases. The dual diffraction engineering scheme offers a clear physical picture for beamforming and greatly simplifies the device structure, with additional merits of large aperture and low power consumption.

Photoacoustic endomicroscopy combined with ultrasound (PAEM-US) has been a long-standing expectation for gastrointestinal tumor examination. Here, we introduce a prototype disposable PAEM-US catheter and corresponding power interface unit, featuring catheter switchability, self-internal three-dimensional scanning, and system repeatability for gastrointestinal endoscopy. By utilizing high-fluence relays, cascade insertion loss of the optical waveguide is minimized to 0.6 dB with a high performance of power resistance, and a focus-customizable acousto-optic coaxial probe is designed for high-sensitivity optical-resolution photoacoustic imaging. Imaging capability was demonstrated with in vivo anatomical imaging at 30 frames per second. Imaging results showed co-registered microscopic visualization of the microvascular and stratification of the rat colorectum with lateral resolution of 18 μm and axial resolution of 63 μm, holding great potential in the clinical detection of gastrointestinal diseases.

As Moore’s law has reached its limits, it is becoming increasingly difficult for traditional computing architectures to meet the demands of continued growth in computing power. Photonic neural computing has become a promising approach to overcome the von Neuman bottleneck. However, while photonic neural networks are good at linear computing, it is difficult to achieve nonlinear computing. Here, we propose and experimentally demonstrate a coherent photonic spiking neural network consisting of Mach–Zehnder modulators (MZMs) as the synapse and an integrated quantum-well Fabry–Perot laser with a saturable absorber (FP-SA) as the photonic spiking neuron. Both linear computation and nonlinear computation are realized in the experiment. In such a coherent architecture, two presynaptic signals are modulated and weighted with two intensity modulation MZMs through the same optical carrier. The nonlinear neuron-like dynamics including temporal integration, threshold, and refractory period are successfully demonstrated. Besides, the effects of frequency detuning on the nonlinear neuron-like dynamics are also explored, and the frequency detuning condition is revealed. The proposed hardware architecture plays a foundational role in constructing a large-scale coherent photonic spiking neural network.

The spectral domain interferometer (SDI) has been widely used in dimensional metrology. Depending on the nature of the SDI, both wider spectral bandwidth and narrower linewidth of the light source are paradoxically required to achieve better resolution and longer measurable distances. From this perspective, a broadband frequency comb with a repetition rate high enough to be spectrally resolved can be an ideal light source for SDIs. In this paper, we propose and implement a broadband electro-optic frequency comb to realize a comb-mode resolved SDI. The proposed electro-optic frequency comb was designed with an optically recirculating loop to provide a broadband spectrum, which has a repetition rate of 17.5 GHz and a spectral range of 35 nm. In a preliminary test, we demonstrated absolute distance measurements with sub-100 nm repeatability. Because of these advantages, we believe this electro-optic frequency comb can open up new possibilities for SDIs.

With the development of controllable quantum systems, fast and practical characterization of multi-qubit gates has become essential for building high-fidelity quantum computing devices. The usual way to fulfill this requirement via randomized benchmarking demands complicated implementation of numerous multi-qubit twirling gates. How to efficiently and reliably estimate the fidelity of a quantum process remains an open problem. This work thus proposes a character-cycle benchmarking protocol and a character-average benchmarking protocol using only local twirling gates to estimate the process fidelity of an individual multi-qubit operation. Our protocols were able to characterize a large class of quantum gates including and beyond the Clifford group via the local gauge transformation, which forms a universal gate set for quantum computing. We demonstrated numerically our protocols for a non-Clifford gate—controlled-(TX) and a Clifford gate—five-qubit quantum error-correcting encoding circuit. The numerical results show that our protocols can efficiently and reliably characterize the gate process fidelities. Compared with the cross-entropy benchmarking, the simulation results show that the character-average benchmarking achieves three orders of magnitude improvements in terms of sampling complexity.

The interest in dynamic modulation of light by ultra-thin materials exhibiting insulator–metal phase transition, such as VO2, has rapidly grown due to the myriad industrial applications, including smart windows and optical limiters. However, for applications in the telecommunication spectral band, the light modulation through a thin VO2 film is low due to the presence of strong material loss. Here, we demonstrate tailored nanostructuring of VO2 to dramatically enhance its transmission modulation, reaching a value as high as 0.73, which is 2 times larger than the previous modulation achieved. The resulting designs, including free-topology optimization, demonstrate the fundamental limit in acquiring the desired optical performance, including achieving positive or negative transmission contrast. Our results on nanophotonic management of lossy nanostructured films open new opportunities for applications of VO2 metasurfaces.