Optical aberrations caused by the scattering of biological tissues limit the imaging performance of optical systems. We have explored a fluorescence confocal imaging technique that utilizes wavefront shaping indirectly, operating in the near-infrared II range. First, we synthesized a highly efficient fluorescent probe in the near-infrared II range, where reducing the scattering of biological tissue can enable high-contrast biopsy imaging. Second, the study investigate the adaptive optical method, utilizing indirect wavefront measurement. The indirect wavefront shaping technology was applied to the laser scanning confocal system, enabling the measurement and compensation of optical aberrations caused by biological tissues, and obtaining imaging of biological tissues with a high signal-to-noise ratio. Finally, an indirect wavefront shaping-based near-infrared II fluorescence confocal imaging system was deployed and relavant experiments were conducted. The experimental outcomes reveal that the system effectively compensates for the aberrations induced by air plates, scattering media and mouse skull, and increases the final signal intensity to 1.47, 1.95 and 2.85 times, respectively. As a result, the final imaging quality is significantly enhanced.

Small-size parts have a small surface area and complex structure. The traditional mark splicing method is easy to destroy the measurement data of the surface, and it is not allowed to paste marks on the surface of some parts. The feature splicing method is not applicable to parts with repetitive features. We propose a scanning measurement method without marks based on mechanical splicing. First, the camera calibration method based on photogrammetry is used to reconstruct the high precision 3D coordinates of the target on the calibration board. By tracking the position of the encoded target, the rotation matrix corresponding to different angles of the turntable is established, and the direction vector of the rotation axis and the fixed point coordinates on the axis are solved. Then the synchronous calibration of the rotation axis and the camera is completed. Second, based on the accurate calibration of poses of two rotation axes, the rotation mosaic matrix is constructed by using the turntable angle to realize the rough registration of multi-view point clouds. Finally, based on the Normal Iterative Closest Point (NICP) algorithm, the fine registration of the point clouds is completed. Experimental results show that the angle error between the two rotation axes calibrated by the target tracking method is 0.023° lower than that of the traditional standard ball fitting method. After calibration, the average size error of the standard ball is less than 0.012 mm. In the automatic measurement of small-size parts, the point cloud splicing effect of the mechanical splicing method after fine registration is similar to that of the mark splicing method, and the splicing stability is higher. The mechanical splicing method is suitable for the 3D topography measurement of small-size parts where the marks cannot be pasted.

Narrow linewidth fiber lasers, which are based on a oscillator seed source with multiple longitudinal-mode, offer substantial advantages for engineering applications and space-limited platform payloads. Additionally, they are considered ideal sub-modules for high-power spectral combinations. The time domain of this type of seed is unstable due to the self-pulse effect, causing significant spectral broadening and stimulated Raman scattering effects during the amplification process, which limits their further application. This paper introduces four commonly used narrow linewidth seeds. The mechanism and suppression methods of the self-pulse effect in multi-longitudinal mode oscillator seeds were analyzed. Critical technologies essential for the optimization and relevant progress of the multi-longitudinal-mode oscillator seed source and amplifier stages are discussed in detail. A future development outlook is also presented. This paper serves as a useful reference for the design of narrow linewidth fiber lasers based on the multi-longitudinal-mode oscillator seed source.

Remote sensing satellites play an crucial role in both national defense and civil exploration. However, the imaging quality of high-resolution remote sensing satellites is significantly affected by atmospheric turbulence. This paper focuses on the impact of camera aperture, satellite orbit altitude and atmospheric turbulence intensity on the imaging quality of space cameras during remote sensing satellite Earth detection. Firstly, the turbulence wavefront simulation method based on the spherical wave model and Kolmogorov turbulence theory is analyzed. Subsequently, the disturbed wavefront, impacted by the camera aperture, satellite orbit height and atmospheric turbulence intensity, is analyzed, and a universal formula is derived. In addition, an equation for imaging resolution, taking into account aperture, satellite orbit height and atmospheric coherence length, is developed. Finally, this study examines the effect of atmospheric turbulence on the modulation transfer function (MTF). The analysis focuses on the relative error of MTF concerning camera aperture, satellite orbit height and atmospheric coherence length, with reference to an MTF value of 0.15. This study provides a theoretical basis for designing, analyzing, and assessing high-resolution remote sensing satellites.

The movement of objects is everywhere in nature. With the rapid development of smart vehicle and 6G mobile communications, the demand for highly Integrated Sensing and Communication (ISAC) devices with communication and motion sensing is increasing. Based on the coexistence of luminescence and detection characteristics of GaN multiple quantum wells, an integrated optoelectronic chip based on the epitaxial GaN multiple quantum wells material on sapphire substrate with sensitive motion detection and visible light communication. The transmitter of the optoelectronic chip transmits a visible light signal in blue band to the moving target object. The visible light signal modulated by the motion of the target object is reflected back to the receiver of the chip to stimulate the changing photocurrent. By analyzing the changing photocurrent, the motion of the target object rotating at different speeds can be detected. The change period of the photocurrent curve is consistent with the rotation period of the target object. We also study the optoelectronic characteristics and the visible light communication performance of the optoelectronic chip. This chip can be used as transceiver terminal of visible light communication system and can also process and transmit the motion detection signals collected by the chip. The optoelectronic chip based on GaN multiple quantum wells materials is a highly integrated ISAC terminal device with application value.

Taking advantage of the high absorption coefficient, excellent photoelectric properties, and high carrier mobility of Single-Walled Carbon NanoTubes (SWCNTs), high-performance, transparent, all-carbon Field-Effect Transistor (FET) photodetector has been constructed with a high transmittance more than 80% in the visible light band, in which semiconducting SWCNT (sc-SWCNT)/fullerene (C₆₀) heterojunctions as the channel materials, patterned metallic SWCNT film as source/drain electrodes, graphene oxide (GO) as the dielectric layer, and Indium Tin Oxide (ITO) as a transparent gate electrode. The electrical test results show that the photodetector exhibits a strong gate-tunable characteristics, and achieves a broadband spectral response from 405 to 1064 nm in the visible-near infrared spectral region. Under 940 nm illumination with a light density of 5 mW/cm², the maximum photoelectric responsivity of 18.55 A/W and a specific detectivity of 5.35×10¹¹ Jones can be achieved.

A ground-based Doppler Asymmetric Spatial Heterodyne (DASH) interferometer with a high Signal-to-Noise Ratio (SNR) and large etendue (AΩ) with thermal compensation was developed to detect wind field information in the middle atmosphere. The detailed parameters and index of the DASH interferometer were developed for the 557.7 nm oxygen airglow spectral line. The system was designed with an expanded Field Of View (FOV) and thermal compensation. The half-FOV angle reached 2.815°, the etendue was 0.09525 cm²sr, and the system’s SNR was approximately 113.75. Through the thermal compensation design, the final optical path difference with temperature variation (dΔd₀/dT) was only 2.224×10⁻⁷mm/°C. The optical system was designed and optimized according to the corresponding parameters. Image-side telecentric and bilateral telecentric optical system structures were used in the entrance optics and exit optics, respectively, and parameters such as telecentricity and distortion met the detection requirements. To verify the design results, a ground-based DASH interferometer experimental platform was constructed, and indoor and outdoor ground-based experiments were conducted. In the final experiment, clear interference fringes were obtained, which proves that the system design results of the DASH interferometer are reasonable, and the system’s SNR and etendue meet the detection requirements.

To improve the extinction ratio of a polarization beam splitter, we propose a dual-slot ultra-compact polarization splitter (PBS) consisting of a hybrid plasma Horizontal Slot Waveguide (HSW) and a silicon nitride hybrid Vertical Slot Waveguide (VSW). The coating material is silicon dioxide, which can prevent the oxidation of the mixed plasma and also facilitate integration with other devices. The mode characteristics of the HSW and VSW are simulated by using the Finite Element Method (FEM). At suitable HSW and VSW widths, the TE polarization modes in HSW and VSW are phase-matched, while the TM polarization modes are phase mismatched. Therefore, the TE mode in an HSW waveguide is strongly coupled with a VSW waveguide by adopting a dual-slot, while the TM mode directly passes through the HSW waveguide. The results show that PBS achieves an Extinction Ratio (ER) of 35.1 dB and an Insertion Loss (IL) of 0.34 dB for the TE mode at 1.55 μm. For the TM mode, PBS reached 40.9 dB for ER and 2.65 dB for IL. The proposed PBS is designed with 100 nm bandwidth, high ER, and low IL, which can be suitable for photonic integrated circuits (PICs).

The decoherence of temporal quantum correlation is explored in a voltage-controlled quantum dots molecule coupled to a cavity. The temporal correlation in the optoelectronic hybrid system is studied based on Leggett-Garg inequalities. The inequality violations can be interpreted as the existence of temporal quantum correlation during dynamical evolution. The temporal quantum correlation is enhanced by its electron tunnel’s strength and cavity frequency detuning. It is found that there is no temporal quantum correlation in the regions where the values of spatial quantum correlation are zero and the maximal violations occur in conditions with high values of quantum correlation. In contrast, the spatial quantum coherence can still exsit when the value of temporal quantum correlation is zero. The method of open quantum system dynamic is used to study the effect of reservoir memory on temporal quantum correlation. The temporal quantum correlation can be suppressed due to the spontaneous decay of the quantum dots and cavity leakage. These results are helpful for quantum information processing technology in hybrid quantum systems.

An all-optical half-adder is designed by combining the nonlinear effect and linear interference effect of photonic crystals. By dividing the light source into two parts equally, the half adder AND gate and XOR gate are designed separately. The nonlinear effect is used to realize the AND gate with high contrast, and the linear interference effect is used to realize the XOR logic, so that the overall response speed of the device is improved. In this design structure, the device only has threshold requirements for the signal light source power. When the signal power is greater than 51.4 mW/μm², it has stable output and strong anti-interference ability. The designed contrast of the half adder carry output port is 20.69 dB, and the output port contrast is 20.13 dB. The data transfer rate is 0.75 Tbits/s and the occupied area is 623 μm².

The propagation loss of a waveguide is a key indicator to evaluate the performance of an integrated optical platform. The commonly used cut-back method for measuring propagation loss requires the introduction of the spiral test structure. In order to remove bending loss, the bending radius is usually designed to be larger but this consequently has a larger footprint. In this paper, we suggested a method to simultaneously measure the propagation loss and bending loss of waveguides with a cut-back structure. According to simulations, the bending loss can be exponentially fitted with the bending radius, which can be further simplified as linear fitting between the natural logarithm of the bending loss and bending radius. A genetic algorithm was used to fit the insertion loss curve of the cut-back structure and the propagation losses and bending loss were calculated. With this method, we measured a cut-back structure of lithium niobate waveguide and got a propagation loss of 0.558 dB/cm and a bending loss of 0.698 dB/90° at a radius of 100 μm and wavelength of 1550 nm. Using this method, we can simultaneously measure waveguide propagation loss and bending loss while mitigating the footprint.

To achieve high-precision and stable tracking performance, a novel disturbance observer for the photoelectric platform based on dual velocity loops is designed. This method aims to minimize the impact of internal friction torque, external carrier disturbances and sensor noise, thereby enhancing the dynamic response performance of the system. Firstly, the mathematical model of double speed-loop is established. By analyzing the signal spectrum and response performance of various sensors, we have chosen the circular grating sensor with low noise and short delay to replace the traditional measuring machine for closing the inner speed loop. Moreover, the Fiber Optic Gyro (FOG) is utilized for the feedback device of the outer speed loop. Then, a disturbance observer is designed based on the gyro speed signal to observe the disturbance compensation residual in the inner speed loop and the outer carrier disturbance signal, while the feed-forward compensation is performed. The experimental results demonstrate that the double speed loop observer control method can reduce the system regulation time to 45% of the original. When subjected to sinusoidal disturbance signals with varying amplitudes (0.25° to 2°) and frequencies (0.25 Hz to 2 Hz), this method effectively improves the system's ability to suppress disturbances and increases the isolation degree from the initial 20.9 dB to 30 dB. The disturbance observer with double speed loops meets the system requirements of rapid response, stable tracking, high precision and strong anti-disturbance ability of the photoelectric tracking platform.

To improve the field of view of the near-to-eye display imaging system by using the optical waveguide scheme, we propose a method of multiplexing interference fringes with equal periods and variable inclination angles to expand the angular bandwidth of the volume holographic grating for augmented reality glasses coupling element. With this method, the range of incident angles after the expansion mathes the Bragg condition, and the influence of the periodic change on the diffraction angle of the incident light is eliminated, thereby improving the angular response range of the coupling element volume holographic grating and reducing the stray light introduced by grating diffraction. The rigorous coupled wave analysis theory is used to simulate the volume holographic grating multiplexing three interference fringes with equal period and variable inclination angles. Under the TE and TM polarization states at the wavelength of 530 nm, the angular bandwidth of the multiplexed volume holographic grating is 3.6° and 3.3°, respectively. The angular bandwidth of the multiplexed volume holographic grating is twice as large as that of the volume holographic grating recorded with a single interference fringe. This method is expected to break the limitation of the volume holographic grating material on the angular bandwidth of the grating, and can be used to expand the field of view of the near-to-eye display imaging system to achieve lightweight, high-efficiency, large-field-of-view, and low-stray-light augmented reality glasses.

Wet transferring two-dimension (2D) material to a semiconductor substrate is a common method to prepare a hetero-junction photodetector. When preparing to wet transfer a hetero-junction, different preparation details have significant effects on the properties of the hetero-junction formed by the 2D materials and semiconductors. In this paper, a series of identical Gr/Si hetero-junction devices were prepared by the wet transfer method and the relationship between its preparation technique and the voltage-current characteristics was studied in detail. The experimental results show that the gradient drying process can significantly reduce the dark current of the Gr/Si hetero-junction photodetector, the optimal drying temperature peak is 170 °C, and the leakage current basically no longer changes above 170 °C. The surface impurities and residual water in the inter-layer of Gr/Si van der Waals hetero-junction has a significant effect on the leakage current of the hetero-junction. The selective etching and annealing process of a Gr/Si van der Waals hetero-junction can also greatly reduce the leakage current. Therefore, a suitable drying process, selective etching process and annealing process are each necessary in the preparation of a Gr/Si hetero-junction photodetector. These results can give reference to the fabrication of two-dimensional material hetero-junction devices by the wet transfer method.

For data communication between single wavelength laser communication terminals, good isolation between signal transmission and reception is the key to establishing duplex bidirectional laser communication. In this paper, with respect to the transmission and reception scheme of a single laser wavelength laser communication terminal and its overall communication performance, the influence of the surface roughness and contamination level of key components on the isolation performance of the laser communication terminal is analyzed. The model parameters are derived from Harvey model and ABg model, and the designed scheme is analyzed using TracePro software. When the surface roughness or contamination level of λ/2 wave plate, λ/4 wave plate and optical antenna structure in the signal transmission channel is improved, the backscattering caused by these elements will reduce the isolation performance in the signal transmission channel. At the same time, the measurement result of laser communication terminal isolation is 77.86 dB, which is basically consistent with the software simulation result of 78.35 dB. This can be applied in laser communication system.

Ammonia emission will cause harm to the environment and human health, so it is particularly important that the ammonia concentrations are measured with high precision. Off-Axis Integrating Cavity Output Spectroscopy (OA-ICOS), which has the advantages of high sensitivity and high response speed, is used to design a high-precision ammonia detection device. The gas absorption cell is composed of two high reflection mirrors with a reflectivity of 99.99%, and the base length of the optical resonator is 30 cm. Finally, an optical path of nearly 3000 m was realized. The Distributed Feedback Laser (DFB) with a central wavelength of 1528 nm is tuned to 6548.611 cm⁻¹ and 6548.798 cm⁻¹. The concentration of NH₃ is changed from 1×10 ⁻⁵ to 5×10⁻⁵ and is detected under an atmospheric pressure of 18.6 kPa at room temperature. The measurement results show that the linear fit R² between NH₃ concentration and signal amplitude can reach 0.99979. The Allan variance is used to analyze the experimental data, and the minimum detection limit of the system can reach 7×10⁻⁹ at 103 s. The experimental results show that the detection device has good stability and high sensitivity, meets the demand for the high-precision detection of ammonia gas, and also provides technical experience for the domestic independent research and development of high-precision detection equipment for trace gases.

Traditional analytical theory design scheme faces problems, such as high computational complexity, limited analytical solution, and high time-consumption. To cambat these issues, based on the design of traditional optical devices, a scheme for designing an optical power splitter with adjustable split ratio according to the reverse design method is proposed. In a compact region of 1.92 μm×1.92 μm, Ge₂Sb₂Se₄Te₁(GSST) is introduced to change the refractive index distribution of the device. The direct binary search algorithm is utilized to search the optimal state distribution of GSST in crystalline and amorphous states. A T-shaped optical power splitter with adjustable split ratio is designed and implemented for the same device structure. The initial structure, split ratio, phase change material region state distribution, manufacturing tolerance, and light field distribution of the device are simulated and analyzed. The results show the minimum relative errors of the designed optical power splitters with three splitting ratios of 1∶1, 1.5∶1 and 2∶1 between wavelengths 1530 nm and 1560 nm are 0.004%, 0.14% and 0.22%, respectively. The maximum fluctuations of the transmission curve in the manufacturing tolerance range are 0.95 dB, 1.21 dB and 1.18 dB, respectively. The splitter has a compact structure and great potential for applications in optical communication and information processing.

We develop a fiber Bragg grating accelerometer based on a bearing and flexure hinge for the measurement of medium-high frequency vibration signals. The mathematical model between its natural frequency and sensitivity and structural parameters is derived based on a mechanical model, and the structural design is optimized based on the theoretical analysis results. With these prerequisites, the sensor was fabricated. Ultimately, its dynamic characteristics are validated using a finite element simulation and vibration experiment. The results show that both its operating frequency range and acceleration sensitivity are 10−1200 Hz and 17.25 pm/g. In addition, this proposed sensor has some advantages such as an error of less than 0.3 g, a good linearity of greater than 0.99, a repeatability error of 2.33%, and it is free of temperature.

In order to solve the problem of aberration detection failure caused by double-layer reflected light of the fundus retina in standard animal model mouse during wavefront detection, a mouse eye aberration measurement technique combined with optical mask modulation was proposed to improve the accuracy of wavefront aberration measurement. First, according to the key parameters of mouse retina, we established the optical system model of mouse eye wavefront aberration detection and performed optical simulations. Then, the effects of optical masks with different apertures on the reflection beam of the non-target layer of the retina were analyzed and compared, and then the parameters of the optical mask and the experimental plan were determined. Finally, the wave front aberration detection system of the mouse eye was established, and the wavefront aberration of the mouse eye was measured in vivo. The experimental results show that the optical mask with 0.5 mm aperture can reduce the root mean square error of mouse eye wavefront aberration measurement by 74.9%, which is similar to the shielding effect of non-target layer reflected in 80% of the theoretical simulation. It can effectively block the reflected light from the non-target layer of the mouse retina, improve the detection accuracy of the wavefront aberration of the mouse eye, and lay a foundation for the further realization of high-resolution imaging of the mouse eye.

In the space gravitational wave detection Taiji mission, a heterodyne laser interferometer is used to detect gravitational wave signals in the middle and low frequency bands. In the Taiji mission, the laser interferometry system is composed of multi-channel interferometers, which involves the phase acquisition and readout of multiple sets of the interference signals. Therefore, the multi-channel phase measurement system is one of the key core technologies of the space laser interferometry. In this paper, a multi-channel phase measurement system is proposed, designed and tested based on the requirements of the Taiji mission and its ground-based laser interferometry experiments. First, the hardware and software design of the multi-channel phase measurement system is given, including hardware architecture design, phase measurement algorithm based on digital phase-locked loop and its implementation on FPGA, software architecture design, etc. Second, a time-domain functional tests of the multi-channel phasemeter are performed, which includes the phase accuracy and linearity. The results show that the dynamic and static phase linearity and accuracy of the multi-channel phase measurement system under different working conditions are good. Finally, the frequency domain noise tests of different channels at different frequencies and different amplitudes are carried out. The results show that the phase noise level of the multi-channel phase meter designed in this paper is better than \begin{document}$ 2{\text{π}} {\text{µ}} {\rm{rad}}/\sqrt{{\rm{Hz}}} $\end{document} in the frequency band of 0.1 mHz−1 Hz. There is good consistency between different channels, and the phase noise introduced by channel differences or ADC chip differences is negligible in the target frequency band. For any interference signal with a frequency between 5−25 MHz, the phasemeter can meet the requirements within the target frequency band. Therefore, the multi-channel phase measurement system meets the requirements of space gravitational wave detection and ground-based interference experiments. At the same time, the research results of this paper also provide an experimental basis for expanding the phase measurement system with more channels in the future.

In the spaceborne gravitational wave interferometric detection, the problem of stray light has received long-term attention. The laser light emitted by the local interferometer produces backward coherent stray light when passing the telescope while the radiation from space that is incident to the spacecraft produces forward incoherent stray light. Forward incoherent stray light has received less attention at this point, but it is a necessary factor of gravitational-wave telescope design. Therefore, this paper studies stray light produced by space gravitational wave telescopes in orbit. First, the annual solar angle is calculated according to the orbital data of the three-star satellite formation of the Taiji Project, and the solar radiation around the 1064 nm band is evaluated. The baffle shadowing function is derived, which satisfies the requirement for the baffle design. The telescope is then modeled optically and mechanically and scatter measurements are conducted for critical optical components. Finally, the stray light reaching the pupil of the telescope is determined based on the energy of the incident sunlight. The results show that when the angle between the incident light and the optical axis is 60°, the stray radiation at the exit pupil is 3.9×10⁻¹² W, and the corresponding point source transmittance is 8.7×10⁻⁹ which meets the requirement for space gravitational waves to detect extremely low levels of stray light.

To solve the high false alarms caused by complex background clutters in infrared small-target detection, a novel detection method based on \begin{document}${L_{1 - 2}}$\end{document} spatial-temporal total variation regularization is proposed. First, the input infrared image sequence is transformed into a Spatial-Temporal Infrared Patch-Tensor (STIPT) structure. This step can associate the spatial and temporal information by using the high dimensional data structures in the tensor domain. Then, weighted Schatten p-norm and \begin{document}${L_{1 - 2}}$\end{document} spatial-temporal total variation regularization are incorporated to recover the low-rank background component to preserve the strong edges and corners, which can improve the accuracy of sparse target component recovery. Finally, the STIPT structure can be transformed into an infrared image sequence by the inverse operator, and an adaptive threshold segmentation is used to obtain the real target. The method is verified using a contrast test with other five methods, and the experimental results show that the false alarm rate by this method decreases to 71.4%, 71.7%, 68.5%, 74.3% and 20.47% compared with the Maxemeidan, Tophat, LIRDNet, DNANet and WSNMSTIPT algorithms. The time cost also decreased to 42.4%, 82.9% and 28.7% of that of the Maxemeidan, DNANet and WSNMSTIPT. The extensive experimental results demonstrate the superiority of this method in detection performance, which can greatly improve the accuracy and efficiency of target detection with complex background clutters.

In order to better preserve high-frequency information in demosaicing multispectral images, we propose a new demosaicing method for multispectral images based on an improved guided filter. Firstly, the strong correlation between adjacent pixels based on the autoregressive model is constructed, gradually estimates the model parameters at each pixel, and the optimal estimation value is obtained by minimizing the estimation error in the local window, interpolates the sampling dense band G, and generates high-quality guide images. The windowed intrinsic variation coefficient is then introduced into the penalty factor to obtain a weighted guide filter with edge sensing ability and to reconstruct the remaining sparse sampling bands. Finally, the CAVE dataset and the TokyoTech dataset are used for simulation. The experimental results show that compared with the mainstream five-band multispectral image demosaicing method, the peak signal-to-noise ratio and structure similarity of the reconstructed image in the CAVE dataset and the TokyoTech dataset are improved by 3.40%, 2.02%, 1.34%, 0.30% and 6.11%, 5.95%, 2.28%, 1.42%, respectively. The local structure and color information of the original image are also better preserved, and the edge artifacts and noise are reduced.

Vehicle infrared image target detection is an important way of road environment perception for autonomous driving. However, existing vehicle infrared image target detection algorithms have defects, such as low memory utilization, complex calculation and low detection accuracy. In order to solve the above problems, an improved YOLOv5s lightweight target detection algorithm is proposed. Firstly, the C3Ghost and Ghost modules are introduced into the YOLOv5s detection network to reduce network complexity. Secondly, the αIoU loss function is introduced to improve the positioning accuracy of the target and the networks training efficiency. Then, the subsampling rate of the network structure is reduced and the KMeans clustering algorithm is used to optimize the prior anchor size to improve the ability to detect of small targets. Finally, coordinate attention and spatial depth convolution modules are respectively introduced into the Backbone and Neck to further optimize the model and improve the feature extraction of the model. The experimental results show that compared with the original YOLOv5s algorithm, the improved algorithm can compress the model size by 78.1%, reduce the number of parameters and Giga Floating-point Operations Per Second by 84.5% and 40.5% respectively, and improve the mean average precision and detection speed by 4.2% and 10.9%, respectively.

The resolution of optical imaging is limited by the diffraction limit, system detector size and many other factors. To obtain images with richer details and clearer textures, a multi-scale feature attention fusion residual network was proposed. Firstly, shallow features of the image were extracted using a layer of convolution and then the multi-scale features were extracted by a cascade of multi-scale feature extraction units. The local channel attention module is introduced in the multi-scale feature extraction unit to adaptively correct the weights of feature channels and improve the attention to high frequency information. The shallow features and the output of each multi-scale feature extraction unit were used as hierarchical features for global feature fusion reconstruction. Finally, the hight-resolution image was reconstructed by introducing shallow features and multi-level image features using the residual branch. Charbonnier loss was adopted to make the training more stable and converge faster. Comparative experiments on the international benchmark datasets show that the model outperforms most state-of-the-art methods on objective metrics. Especially on the Set5 data set, the PSNR index of the 4× reconstruction result is increased by 0.39 dB, and the SSIM index is increased to 0.8992, and the subjective visual effect of the algorithm is better.

In this paper, an image super-resolution reconstruction multi-scale algorithm based on a residual attention network (SMRAN) is proposed to solve the problems caused by low resolutions, less texture information and blurred details in colorectal endoscopic images. Images from the colorectal polyp endoscope image dataset PolypsSet are selected as the raw data for these experiments. A convolutional network is built to extract the shallow features of the low-resolution image and a Res-Sobel block is designed to enhance its edge features. A multi-scale feature fusion block MEB is designed by introducing convolution kernels of different sizes to adaptively extract image features of different scales and obtain effective image information. The Res-Sobel block and multi-scale feature fusion module block MEB are connected through the residual attention network. Finally, a high-resolution image is reconstructed at the sub-pixel convolution layer. When the amplification factor is ×4, the performance of the proposed algorithm on the test set are as follows: the peak signal-to-noise ratio (PSNR) is 34.25 dB and the structural similarity (SSIM) is 0.8675. Compared with the traditional bicubic interpolation algorithm and commonly used deep learning algorithms such as SRCNN and RCAN, the proposed SMRAN algorithm shows better super-resolution reconstruction results on colorectal endoscopic images.

Miniature head-mounted single-photon fluorescence microscopy is a breakthrough approach for neuroscience research that has emerged in recent years. It can image the neural activity of freely moving vivo animals in real time, providing an unprecedented way to access neural signals and rapidly enhancing the understanding of how the brain works. Driven by the needs of brain science research, there have been many types of miniature head-mounted single-photon fluorescence microscopes, such as high-resolution imaging, wireless recording, 3D imaging, two-region imaging and two-color imaging. In order to have a more comprehensive understanding of this new optical neuroimaging technology, we classify its technologies according to the imaging field of view, introduce the characteristics of different types of micro-head-mounted single-photon fluorescence microscopes reported so far, and focus on the optical system scheme and optical performance parameters used. The advantages and disadvantages of different schemes are analyzed and compared and the future direction of development is described to provide reference for the practical application of brain science researchers.

In order to clarify the cavity design methods of thin-disk multi-pass amplifiers, we summarize the different types of thin-disk multi-pass amplifiers and concludes that there are four fundamental design concepts: (1) 4f relay imaging, (2) resonant cavity design/optical Fourier transform, (3) near-collimated beam transmission, and (4) others. Each amplifier design method is described and the current status of its research is listed in as much detail as possible. By comparing the four types of disk multi-pass amplifiers, it is found that the varying methods have distinct advantages and disadvantages. 4f relay imaging requires a vacuum environment to avoid gas ionization at the focal point, making the mechanics and adjustment more difficult; the resonant cavity design/optical Fourier transform concept multi-pass amplifier has a small spot at the mirrors, making it more suitable for lower energy multi-pass amplifiers; the near collimated beam transmission method has great development potential because it does not require a vacuum environment, but accurately controlling the surface shape of the thin-disk is difficult while the laser is operating. Therefore, from the perspective of laser design, it is necessary to continue to optimize the design of the thin-disk multi-pass amplifier to realize the diversification of application scenarios and the sustainable expansion of output energy.

Polarization imaging, a novel photoelectric detection technology, can simultaneously acquire the contour information and polarization features of a scene. For specific application scenarios, polarization imaging has the excellent ability to distinguish different objects and highlight their outlines. Therefore, polarization imaging has been widely applied in the fields of object detection, underwater imaging, life science, environmental monitoring, 3D imaging, etc. Polarization splitting or the filtering device is the core element in a polarization imaging system. The traditional counterpart suffers from a bulky size, poor optical performance, and being sensitive to external disturbances, and can hardly meet the requirements of a highly integrated, highly functional, and highly stable polarization imaging system. A metasurface is a two-dimensional planar photonic device whose comprising units are arranged quasi-periodically at subwavelength intervals, and can finely regulate the amplitude and phase of the light field in different polarization directions. Polarization devices based on metasurface are featured with compactness, lightweight and multi-degree freedom, offering an original solution to ultracompact polarization imaging systems. Targeted at the field of polarization imaging, this paper illustrates the functional theory, developmental process and future tendency of related metasurfaces. We discuss the challenges and prospect on the future of imaging applications and systematic integrations with metasurfaces.

With the development of industrial manufacturing towards intelligence, precision and integration, on-machine verification of the machining process can provide timely feedback on measurement results, compensate and correct processing parameters, thereby aiding in enhancing machining accuracy and efficiency. Fringe structured light technology is a non-contact measurement method, which has developed rapidly in recent years. It has the characteristics of simple measurement principle, low costs, high measurement accuracy and easy integration, which provides a new solution for on-machine verification. However, the accuracy of structural for on-machine verification is compromised by the convoluted lighting in machining environments and metal parts’ high reflectivity, leading to inaccurate measurements. Applying high dynamic range (HDR) technology to structural light detection enables the measurement of metal parts in complex scenes and reduces the effect of high reflectivity. This paper introduces the measurement principle of structured light and summarizes the challenges of on-machine verification for HDR structured light. Subsequently, this paper provides a comprehensive review of HDR structured light technology. Based on the context of on-machine verification, the HDR technology using hardware equipment and the stripe algorithm are discussed and analyzed, respectively. Following this, different technologies are summarized according to the requirements of on-machine verification. The advantages and disadvantages of various methods are presented, and the applicability of on-machine verification is compared. Finally, the potential applications are analyzed, and the technological prospects will be proposed in combination with the research hotspots of advanced manufacturing technology and precision measurement in recent years.

In order to provide high-quality, intelligent, and healthy lighting sources, a linear dimming hybrid lighting system is constructed using three-primary-color LED light sources. An optimization method for dimming and color mixing is proposed. The light chromaticity and light intensity of the mixed light source are set by color temperature and brightness level, which makes the mixed lighting effect more in line with the demand of "human-centric lighting". In the intelligent optimization of the system, the color temperature is converted into CIE \begin{document}$ {u}'{v}' $\end{document} chromaticity coordinates, and the luminance is converted into brightness to make the optimization calculations more accurate. The system adopts the linear dimming method, effectively preventing the health and safety risks caused by light source flicker, and effectively solves the problem of large linear dimming chromaticity drift with an optimization algorithm. Experiments demonstate that the chromatic stability of the mixed light in the hybrid lighting system is maintained within the first step of the CIE \begin{document}${u}'{v}' $\end{document} chromaticity diagram between a color temperature range of 2000 K to 8000 K. Moreover, the system exhibits imperceptible color deviation over the entire range of light intensity adjustments at the corresponding color temperature. This finding also indicate that linear dimming hold a superior edge over pulse-width dimming method in keeping the optical chromaticity stable. Theoretical exploration and experimental results demonstrate that the mixed lighting system is straightforword and practical, with significant practical value.

In order to meet the requirements of the national synchrotron radiation source, the anisotropic wet-etching technology of small blazed angle monocrystalline silicon grating is studied, and the blazed grating suitable for the medium wave soft X-ray band is prepared. Based on the rigorously coupled wave theory, the structural parameters and process tolerance of the small blazed angle grating are designed. In the crystal alignment process, the crystal orientation of the silicon wafer is determined by ring-preetching, and then the grating mask is aligned with the crystal direction of monocrystalline silicon <111> based on the frequency doubling adjustment method. At the same time, the effect of the photoresist ashing technique and the active agent on the groove quality of the grating is investigated, and the scintillating gratings close to the ideal sawtooth groove shape are successfully prepared by the monocrystalline silicon anisotropic wet etching process. The experimental results show that the blazed angle of the prepared grating is 1°, the linear density is 1200 gr/mm, and the root mean square roughness of the blazed surface is less than 0.5nm. This method can be applied to the fabrication of the medium wave soft X-ray band blazed grating, which can greatly reduce the difficulty and cost of fabrication while achieving high diffraction efficiency.

Phase retrieval and depth estimation are vital to three-dimensional measurement using structured light. Currently, conventional methods used for structured light analysis have limited efficiency and produce unreliable results. To enhance the reconstruction effect of structured light, this paper proposes a hybrid network for structured light phase and depth estimation based on Light Self-Limited Attention (LSLA). Specifically, a CNN-Transformer hybrid module is constructed and integrated into a U-shaped structure to harness the complementary benefits of CNN and Transformer. The network is assessed comparatively with other networks in tasks related to structured light phase estimation and depth estimation. The outcome of the experimental indicates that the suggested network achieves finer detail processing in phase and depth estimation compared to other networks. Specifically, in structured light phase anddepth estimation, its accuracy improves by 31% and 26%, respectively. Therefore, the proposed network improves the accuracy of deep neural networks in the aforementioned areas.

The high mismatch rate of the parallax discontinuity region and the repeated texture region has been a major issue affecting the measurement accuracy of binocular stereo matching. For these reasons, this paper proposes a stereo matching algorithm that utilize multi-feature fusion. Firstly, the weight of neighboring pixels is given using Gaussian weighting method, which optimizes the calculation accuracy of the Sum of Absolute Differences (SAD) algorithm. Based on the Census transformation, the binary chain code technique has been enhanced to fuse the average gray value of neighborhood pixels with the average gray value of gradient image, and then the judgment basis of the left and right image corresponding points is established, and the coding length is optimized. Secondly, an aggregation technique has been developed that combines the cross method and the improved guide filter to redistribute disparity values with the aim of minimizing error. Finally, the initial disparity is obtained by the Winner Take All (WTA) algorithm, and the final disparity results are obtained by the left-right consistency detection method, sub-pixel method, and then a stereo matching algorithm based on the multi-feature SAD Census transform is established. The experimental results show that the average error matching rate of the proposed algorithm is 4.18%, the average error of the 200−900 mm distance is less than 2%, and the maximum error of the actual 3D data measurement is 1.5%, all using the test of the Middlebury dataset. The experimental results verify the effectiveness of the proposed algorithm.

To achieve periodic closed-loop correction of multiple visible wavelength lasers, a laser beam combining system is being designed. This system involves independent monitoring and adjusting of beam pointing and position deviation. First, according to the application requirements of the system, the design indexes of the beam combining system and the overall beam combining scheme are proposed. Then, based on the overall beam combining scheme, we establish the beam control model for the beam combining system. Through numerical simulation experiments, we obtain the solution method for beam control of the beam combining system. The closed-loop beam combining system enables independent monitoring of the unit beam’s pointing and position deviation through the respective beam pointing and position monitoring device. The monitoring results are then used to calculate the control quantity of the beam adjusting device. The independent and efficient adjustment of beam pointing, and position deviation is achieved using a two-dimensional swing mirror and a one-dimensional platform, respectively. Finally, a closed-loop beam combining simulation experimental system with beam monitoring and adjustment device was built using tow laser beams of different wavelengths. The periodic closed-loop beam combining system was verified to have an effective beam combing effect. The experimental results demonstrate that over an extended operational period, both lasers achieve precise beam combining with the reference optical path. Furthermore, the beam combining pointing accuracy is better than ±7μrad, and the positioning accuracy is better than ±0.84 mm. The laser beam combining system developed in this study boast high beam combining accuracy, a fast correction speed, and excellent augmentability for multiple laser beams. Besides, it can accomplish periodic closed-loop beam combining of laser beams, ensuring long-term working stability of the combined laser.

In order to meet the urgent need of developing lightweight and compact space cameras quickly, effectively, and rapidly, a detailed comparative analysis is conducted, including optical system forms and imaging systems. The optical system form of RC+ compensation group is determined, and the imaging system of small F#+micropixel is adopted. Compared with the detailed parameters of the DOVE camera, a lightweight all-aluminum high-resolution camera with a resolution of 3.48m at an orbital altitude of 500km is designed. The overall design results of the camera, its optical and optomechanical structure, imaging electronics, and thermal control are described in detail. The optical design results of the RC+ compensation group of F5.6 are obtained. Using RSA-6061 microcrystalline aluminum alloy as the structural material of the mirror, coupled with an integrated high-rigidity hard aluminum alloy structure, the static (gravity and temperature deformation) simulation analysis results meet the optical design tolerance requirements. The dynamic simulation analysis results show that the first order mode is 302.92 Hz, which has a sufficiently high dynamic stiffness and safety redundancy. The imaging electronics using a 3.2 μm large area array 9 K×7 K detector is designed for low noise miniaturization. Thermal control provided by the satellite platform at a temperature level of 20 °C± 4 °C for the camera. Integration test results show that: (1) The RMS wave aberration of the central field of view is λ/15.6, the wavefront aberration of the five fields of view is better than λ/12.3, which ensures high-quality imaging near the diffraction limit of the camera. The measured optical transfer function at Nyquist frequency is 0.217; (2) The maximum sinusoidal vibration of the camera in three directions is amplified 1.17 times, and the first-order mode of the camera is 295 Hz, with a deviation of 2.61% from the simulation result. The structural stiffness is high and the mechanical stability is good. Under vacuum environment of 10⁻⁴ Pa and three different temperatures of 16 °C, 20 °C and 24 °C, the image is clear and can distinguish the corresponding resolution plate image at Nyquist frequency; (3) The image of 2 km outfield target is good, as well as clear and distinct grayscale image with sharp shadow boundaries. The all-aluminum high-resolution camera with lightweight and compact size achieves 3.48 m resolution at a track height of 500 km,width of 15 km×15 km and a total weight of 2 kg. The structural rigidity and strength test results meet the requirements of space launch scenarios, and can provide theoretical guidance and engineering reference for the design of lightweight and higher-resolution space cameras.

Cytoendoscopy requires continuous amplification with a maximum magnification rate of about 500 times. Due to fiber-optic illumination and stray light, the image has non-uniform illumination that changes with the magnification rate, which affects the observation and evaluation of lesions by doctors. Therefore, this paper proposes an image non-uniform illumination correction algorithm based on the illumination model of cytoendoscopy. According to the principle that image information is composed of illumination and reflection components, the algorithm obtains the illumination component of the image through a convolutional neural network, and realizes non-uniform illumination correction based on the two-dimensional Gamma function. Experiments show that the average gradient of the illumination channel and the discrete entropy of the image are 0.22 and 7.89, respectively, after the non-uniform illumination correction by the proposed method, which is superior to the traditional methods such as adaptive histogram equalization, homophobic filtering, single-scale Retinex and the WSI-FCN algorithm based on deep learning.

Time of Flight (ToF) depth camera is one of the important means to obtain three-dimensional point cloud data, but ToF depth camera is limited by its own hardware and external environment, and its measurement data has certain errors. Aiming at the unsystematic error of ToF depth camera, this paper experimentally verifies that the color, distance, and relative motion of the measured target affect the data obtained by the depth camera, and the error effects are different. Based on the oscillation error and the fact that each signal can be expressed in the form of a Fourier series, a new measurement error model is proposed to correct the error caused by color and distance. For the error caused by relative motion, a three-dimensional motion blur function is established to recover it. Through the numerical analysis of the established calibration model, the residual error of distance and color is less than 4mm, and the error caused by relative motion is less than 0.7mm. The work done in this paper improves the quality of the measurement data of the ToF depth camera, and provides more accurate data support for 3D point cloud reconstruction and other work.

We propose cosh-Pearcey-Gauss beams with uniform polarization, which are mainly modulated by a hyperbolic cosine function (n, Ω) and the angles related to uniform polarization (α, δ). Based on angular spectrum representation and the stationary phase method, the Poynting vector, Spin Angular Momentursl (SAM) and Orbital Angular Momentums (OAMs) in the far zone are studied. The results show that a larger n or Ω in the hyperbolic cosine function can partition the longitudinal Poynting vectors, SAMs and OAMs into more multi-lobed parabolic structures. Different polarizations described by (α, δ) can distinguish their Poynting vectors and angular momentums between the TE and TM terms, though this does not affect the patterns of the whole beam. Furthermore, the weight of the left and right sides of longitudinal Poynting vectors, SAMs and OAMs in TE and TM terms can be modulated by left-handed or right-handed elliptical polarization, respectively. The results in this paper may be useful for information storage and polarization imaging.

In order to meet the ultra-high temperature stability requirements of the inertial sensor ground weak force measurement system, the thermal design of the whole system is carried out. Firstly, the structure of ground weak force measurement system of inertial sensor, heat transfer path of sensitive structure and internal heat source are introduced. Secondly, according to the requirements of the thermal control index of the system, a high-precision thermal control method combining the three-stage thermal control structure and proportional integral differential (PID) control algorithm is proposed to reduce the influence of temperature noise on the detection sensitivity of the inertial sensor. Then, UG/NX software is used to establish the finite element model and carry out the thermal analysis calculation under different working conditions, and the temperature change value of the measurement system in the time domain after equilibrium is (1.2−1.6) ×10⁻⁵ K. Finally, the temperature distribution of the measurement system in the time domain is described in the frequency domain, and the temperature stability results of the inertial sensor sensing structure are obtained. The analysis results show that under the current thermal control measures, the temperature stability of the inertial sensor sensing structure is better than 10⁻⁴ K/Hz^1/2, meeting the requirements of thermal control indicators, and the thermal design scheme is reasonable and feasible.

Constructing the radiation characteristics of space targets is of great significance for the development of space situational awareness technology. In this study, we aim to investigate the infrared radiation characteristics of space targets by developing a simulation program based on the finite element method and unstructured tetrahedral mesh. Through vector coordinate transformation, we calculate the orbit external heat flux received by each surface of the target. By combining the surface material properties and bidirectional reflection distribution function (BRDF), we simulate the temperature and infrared radiation characteristics of each target surface were simulated. Furthermore, we analyze the spectral radiation intensity of the target in the ascending and descending orbital arcs under ground-based detection conditions, taking into account the effects of atmospheric attenuation and background radiation. The results show that, for a three-axis stabilized synchronous orbit satellite with solar panels fixed in the flight direction, the temperature variation range of each surface in the sunlight area and the shadow area is small. The detection effect of the long-wave band of 8~14 μm is better than that of the medium-wave band of 3~5 μm, and the maximum radiation intensity is about 770 W/sr. Ground-based infrared spectrum detection is more affected by the atmosphere, and the detection band must be optimally selected.

The anamorphic optical system has a two-plane symmetry, with different focal lengths in the two symmetry planes. This system can obtain a wider field of view when using sensors of conventional size. This paper proposes a method for designing catadioptric anamorphic optical systems based on their first-order aberration characteristics. A catadioptric anamorphic optical system is designed by using a biconic surface, with a focal length of 500 mm in the XOZ plane and 1000 mm in the YOZ plane. The system’s F-number is 10, and the full field angle is 1°×1°. The mean value of the full field of view MTF of the system at is higher than 0.3 at 80 lp/mm. The overall structure of the system is compact, and the imaging quality is excellent.

Shearography is a non-contact, full-field, and high-precision optical deformation measurement technology. There is a lot of random noise in the acquired speckle fringe pattern caused by environmental factors, which affects the measurement accuracy. The traditional denoising methods easily cause the fringe information to be lost or even damaged while filtering out the noise. To solve this problem, this paper proposes an image denoising method based on the combination of sine and cosine transform and bitonic filtering. In this method, the phase fringe image is firstly obtained by sine and cosine transform. Secondly, the two images are denoised by the bitonic filtering method respectively. Finally, the filtered two images are merged into the final phase fringe image. Experimental results show that for the filtered phase pattern, the speckle suppression index is 0.999 and the average retention index is 2.995, which prove that the proposed method can improve the quality of the phase pattern better than the traditional denoising method, and can preserve the details and edge information of the phase fringes to a large extent.

The development of smart cars puts forward higher requirements for intelligent light systems with projection functions. Compared with the vehicle light system using traditional technology, the vehicle light system using Micro LED projection display technology can make the system structure simpler and the light utilization rate higher, which is more conducive to miniaturization, energy saving and integration because of its good self-luminous characteristics. In this paper, we propose a light projection scheme based on Micro LED array, design a 200×150 white Micro LED array with a pixel size of 80 μm×80 μm as the display light source and the field of view angle is 16°×34° object image tilt, and optimize the tilt angle and optical system structure of the object surface. In addition, the reverse distortion processing method and the pixel grayscale modulation method are used to solve the problems of keystone distortion and illumination uniformity of the projected image of the vehicle light, and a projection experimental platform is built to verify the image correction method. The experimental results show that the keystone distortion coefficients of the corrected images in the x and y directions decrease from 0.0932 and 0.3680 to 0.0835 and 0.0373, respectively, and the uniformity of the illumination of the image surface increases from 83.2% to 93.2%. In this paper, the optimization design of the oblique projection vehicle lighting system based on Micro LED and its image correction method are used to realize the projection of the vehicle light with high luminous efficiency and low distortion.

With the increasingly crowded space resources in low orbit, space situational awareness is an important support for the normal operation of space assets, and optical observation is one of the important means. In this paper, the solar radiation, the earth radiation and the earth albedo radiation received by the space target are simulated by Monte Carlo simulation method, and the simulation program is written based on the unstructured tetrahedral grid, and the calculation results are compared and verified. Furthermore, for the orbit external heat flow received by the sun-synchronous orbit satellite, the grid with solar panels is used to analyze the orbit heat flow received by each surface with or without occlusion. The results show that the average heat flow value of -Y surface decreases by 53.79 W/m² after considering occlusion in the ground mode. The average surface heat flow value of +Y-Z side panel decreased by 32.05 W/m². The temperature characteristics of each surface are given based on the properties of surface materials, and the accuracy of the calculation is verified by combining with the on-orbit telemetry data of the solar panel temperature. Finally, the infrared radiation intensity in each direction of the two modes is calculated. The results show that the influence of heat flow on the surface is different under different observation modes. The temperature of the surface varies with time under the earth mode, while the heat flow on the surface is stable under the sun mode. Under the two modes, the temperature of the solar panel is higher, the radiation intensity is larger, and it has obvious infrared characteristics, which is convenient to carry out infrared observation.

Ultraviolet lasers play an important role in the study of ultraviolet resonance Raman spectroscopy. The Raman resonant Raman effect enhance Raman signals and reduces the detection limit of Raman measurement. This paper focuses on the study of an all-solid-state deep-ultraviolet laser with an output wavelength of 228 nm. The laser uses Nd:YVO₄ as a gain medium and electro-optic q-switched cavity dumping technique to achieve a fundamental frequency output of 914 nm in pulse widths of several nanoseconds. Then, the second-harmonic generation is achieved by LiB₃O₅(LBO), and the fourth-harmonic 228 nm UV laser is obtained by beta-barium-borate (BBO). The variation of fundamental and second harmonic laser power at different repetition rates is investigate. The average power of Nd:YVO₄ is saturated and decreases with increased repetition rate due to the low gain at 914 nm. The output efficiency of UV laser is optimized by adjusting the focus lens. At the pump power of 30 W, the highest average power of a 228nm UV laser is 84 mW at 10 kHz. The repetition rate of UV laser is continuously adjustable within the range of 5 kHz−25 kHz, and the pulse width is maintained at 2.8 to 2.9 ns which meets the application requirements in the field of UV spectroscopy detection technology.

The objective of this study is to explore the optimal detection location and the best prediction model of the brix value of Yongquan honey tangerines, which can provide a theoretical basis for the brix measurement and classification of honey tangerines. With the wavelength range of 390.2−981.3 nm hyperspectral imaging system was used to study the best location for detecting the brix of Yongquan honey tangerines, and the spectral information of the calyx, fruit stem, equator and global of Yongquan honey tangerines were combined with their brix values of corresponding parts to establish its prediction model. The original spectra from the different locations were pre-processed by standard normal variance transformation (SNV), multiple scattering correction (MSC), baseline calibration (Baseline) and convolutional smoothing (SG), respectively, and the partial least squares regression (PLSR) and least squares support vector machine (LSSVM) models were established based on the pre-processed spectral data. The best pre-processing methods for different parts of the honey tangerine were found, and the spectral data obtained from these pre-processing methods were analyzed using the competitive adaptive re-weighting algorithm (CARS) and uninformative variable elimination (UVE) to identify characteristic wavelengths. Finally, the PLSR and LSSVM models were established and compared based on the selected spectral data. The results show that the global MSC-CARS-LSSVM model demonstrates the most accurate prediction performance, with a correlation coefficient of Rp=0.955 and an RMSEP value of 0.395. Alternatively, the SNV-PLSR model of the equatorial location of honey tangerines was found to be the next more effective, with a correlation coefficient of Rp=0.936, and an RMSEP value of 0.37. The correlation coefficients of the two prediction models are similar, the equatorial location can be used as the optimal location for measuring the brix of honey tangerines. This study demonstrates that the use of varying segments of the orange impacts the accuracy of prediction models. Identifying the optimal location and prediction model can provide a theoretical basis for classifying oranges for brix testing.

Conventional imaging spectrometers generally have low variable power, which is not conducive to the extended application of large-field, long-slit, multi-channel optical systems. In space remote sensing, the radiation energy of the ultraviolet band is low, which requires the imaging spectrometer to have a smaller F-number. In order to meet the requirement of detecting small F-number of high variable power and high spectral resolution imaging spectrometer, a high spectral resolution Offner UV imaging spectrometer with high variable power is designed in this paper. An improved Offner structure with light and small size is adopted in the rear beam splitting system of the imaging spectrometer. Based on the requirements of variable power ratio and small F-number of the imaging spectrometer, the initial Offner structure parameters are derived theoretically. A meniscus lens is inserted in front of the image to increase the degree of freedom of optimization of the system and improve the imaging quality of the system. The resulting imaging spectrometer works in the 270~300 nm band with a long slit of 40 mm, a spectral resolution better than 0.6 nm, a spectral sampling of 0.15nm, the system variable power ratio less than 0.22, and an F number less than 2. the system modulation transfer function (MTF) is better than 0.9 at a cutoff frequency of 14 lp/mm, and the root mean square radius (RMS) of each field of view in each band is less than 12 μm. This study provides a design scheme for the UV-band hyperspectral detection imaging spectrometer with small F-number and high variable power.

Narcissus refers to the phenomenon in an infrared system where a cold imaging sensor can "see" its own reflected image by the reflection of the frontal optical surfaces. Control of narcissus is one of the important requirements in the design of the infrared imaging system. A cooled medium-wave infrared imaging system with Cassegrain reflection structure is designed and analyzed to obtain the optical surfaces with serious narcissus. In addition, the narcissus is reduced by Zemax, and the optimization of the system transfer function MTF is taken into account while the narcissus is controlled. The optimized medium-wave infrared imaging system is compared through NARCISSUS macro (narcissus analysis macro), Tracepro modeling software and actual imaging, and it was found that the narcissus induced equivalent temperature difference (NITD) of the detector image surface decrease from 1.0484 K to 0.1576 K. The energy and size of the narcissus spot did not show marked change during the focusing of the system. The optimized optical structure effectively controls the narcissus of the system.

To improve the detection efficiency of deep ultraviolet laser for semiconductor detection, it is imperative to develop 257 nm deep ultraviolet picosecond laser with high power and high repetition. In this study, a 257 nm deep ultraviolet laser was experimentally investigated based on photonic fiber amplifier and extra-cavity frequency quadrupling. The seed source uses a fiber laser with a central wavelength of 1030 nm and a pulse width of 50ps, delivering a power output of 20 mW and a repetition frequency of 19.8 MHz. High power 1030 nm fundamental frequency crystal light was obtained through a two-stage ytterbium-doped double cladding (65 μm/275 μm) photonic crystal fiber rod amplification structure, and 257 nm deep ultraviolet laser was generated using double frequency crystal LBO and quadruple frequency crystal BBO. The seed source uses a two-stage photonic crystal fiber amplifier to get a 1030 nm laser output power of 86 W. After the laser focusing system, frequency doubling resulted in a second harmonic output of 47.5 W at 515 nm and a fourth harmonic output of 5.2 W at 257 nm.The fourth harmonic conversion efficiency was 6.05%. The experimental results show that this structure can obtain high power 257 nm deep ultraviolet laser output, providing a novel approach to improve the detection efficiency of semiconductor detection lasers.

Current study on photonic neural networks mainly focuses on improving the performance of single-modal networks, while study on multimodal information processing is lacking. Compared with single-modal networks, multimodal learning utilizes complementary information between modalities. Therefore, multimodal learning can make the representation learned by the model more complete. In this paper, we propose a method that combines photonic neural networks and multimodal fusion techniques. First, a heterogeneous photonic neural network is constructed by combining a photonic convolutional neural network and a photonic artificial neural network, and multimodal data are processed by the heterogeneous photonic neural network. Second, the fusion performance is enhanced by introducing attention mechanism in the fusion stage. Ultimately, the accuracy of task classification is improved. In the MNIST dataset of handwritten digits classification task, the classification accuracy of the heterogeneous photonic neural network fused by the splicing method is 95.75%; the heterogeneous photonic neural network fused by introducing the attention mechanism is classified with an accuracy of 98.31%, which is better than many current advanced single-modal photonic neural networks. Compared with the electronic heterogeneous neural network, the training speed of the model is improved by 1.7 times; compared with the single-modality photonic neural network model, the heterogeneous photonic neural network can make the representation learned by the model more complete, thus effectively improving the classification accuracy of MNIST dataset of handwritten digits.

With excellent optical properties and high carrier mobility, perovskite materials have become highly competitive materials in the field of space solar cells. However, space particle irradiation can change the structure and optical properties of materials, leading to a rapid degradation of device performance. In order to investigate the influence of electron irradiation on the structure and optical properties of CsPbBr₃ nanocrystals, this paper conducted electron irradiation experiments on CsPbBr₃ materials, characterized the microscopic morphology of CsPbBr₃ nanocrystals by high-resolution transmission electron microscopy method. Moreover, this paper investigated the change trend of crystal structure by X-ray diffraction analysis and X-ray photoelectron spectroscopy analysis. The results revealed electron irradiation caused the CsPbBr3 nanocrystals to become rough and significantly reduced in size. The nanocrystal became compact and formed nanocluster under high-dose electron irradiation. Furthermore, the optical properties of CsPbBr₃ materials were characterized using steady-state UV-Vis absorption spectra and photoluminescence spectra. The analysis of lattice expansion-induced bandgap changes after irradiation was performed using first principles calculations. It is demonstrated that electron irradiation deepened the color of nanocrystals and affected the light transmittance of CsPbBr₃ nanocrystalline, thereby enhancing the optical absorption performance of the samples. However, electron irradiation also led to the decomposition of CsPbBr3 nanocrystals, resulting in a significant reduction in luminescence intensity by 53.7%−78.6% after high-dose irradiation. These findings provide valuable data support for the study of spatial radiation damage mechanisms and the application of perovskite nanocrystals.

Optical fiber tweezers have a simple structure, operate flexibly and are small in size, and are widely used in biochemical analysis, life science and other fields. The optical fiber probe with hetero-core structure has natural advantages in near-field evanescent wave optical trapping, core beam coupling transmission, and cross-synergistic applications of microfluidic technology. It can realize cell and subcellular particle collection and transportation and can significantly improve the three-dimensional capture ability and dynamic manipulation level of particles. In this paper, the structural characteristics and application technology research progress of fiber optical tweezers with different core structures are reviewed. Their key technologies such as probe preparation, laser source and coupling mode are listed and compared. The role and development of hetero-core fibers with different structures in fiber optical tweezers are summarized and predicted.

The concentrating solar simulator can obtain high-power convergence of solar radiation spots, which has important applications in the fields of solar thermal power generation and thermochemical research. To obtain uniform solar radiation spots, a free-form condenser design method based on non-imaging optics is proposed, and its design principle and specific method are described. A freeform condenser is designed in comparison to a non-coaxial ellipsoidal condenser with the same containment angle, and the correctness of its method of design is verified by simulation analysis. The simulation results show that when a xenon lamp with a rated power of 6 kW is used as the light source, the single-lamp solar simulator composed of a freeform condenser can produce a spot with an average irradiance of 274.4 kW/m² in the target and a diameter of 60mm. The spot’s unevenness decreases from 18.28% to 5.69% compared with that of a non-coaxial ellipsoidal solar simulator. The seven-lamp solar simulator can produce a spot with an average irradiance of 1.65 MW/m², with a spot unevenness that decreases from 13.19% to 5.49%.

Optical path absorption spectroscopy is an important branch of absorption spectroscopy. In recent years, there has been a proliferation of optical path absorption spectroscopy techniques based on different light source technologies, absorption cavity technologies, and detection methods. As the demands on detection sensitivity and absorption optical path length increased, optical path absorption spectroscopy techniques based on the principle of enhanced absorption emerged, including integrated cavity spectroscopy (ICOS), cavity-enhanced absorption spectroscopy (CEAS) and cavity ring-down spectroscopy (CRDS). Enhanced absorption spectroscopy is advantageous for its high spectral resolution, high sensitivity, fast response time, and portability, but it presently lacks a unified concept and clear classification criteria. This paper compares the development history of absorption spectroscopy techniques and clarifies the concept of their multi-optical path. Based on whether resonant absorption occurs in the absorption cavity, the concept of absorption spectroscopy techniques based on resonance is proposed, and the current research status of resonant absorption spectroscopy techniques is analyzed and summarized, and the applications of this technique in various fields are outlined. Finally, the future development of key technologies in resonance absorption spectroscopy is envisioned.

Underwater polarized light with certain distribution characteristics is formed when sunlight is scattered by the atmosphere and refracted by the surface of the water. The resulting underwater polarization distribution pattern can be used in navigation. In this paper, an air-water model is proposed to calculate the polarization pattern of sky light under varying wave conditions and simulate the underwater polarization distribution pattern under the influence of wave refraction. Distribution images are simulated for underwater polarization’s degree and angle in conditions with calm water, sinusoidal waves and random waves with different solar altitude angles. The results are verified using underwater experiments. The comparison of the polarization distribution pattern under the waves and that under the calm water show that the proposed model can accurately characterize the characteristics of the polarization distribution pattern under typical wave surfaces, providing a theoretical basis for improving the environmental adaptability of underwater polarization navigation under fluctuating water surface conditions.

In order to improve the traditional attitude measurement accuracy of star sensors, interference angle measuring technology can be combined with a traditional star sensor. Based on the centroid positioning technology of traditional star sensors, the light intensity information of star image points is subdivided to break through the accuracy limitation of centroid positioning and form a highly precise interferometric star sensor with a large field of view. In this paper, the factors that restrict the Angle measurement accuracy of interferometer sensors are deeply studied with particular interest given to the influence of interference fringe segmentation error on angle measurement accuracy. Through research and analysis, we conclude that the asymmetry error is not the main factor affecting the angle measurement accuracy of interferometric sensors. When the mismatch error between the Mohr fringe period and the overall optical dimension of the optical wedge array is less than 1%, the single-factor angle measurement error is less than 0.01". For non-orthogonal error between Mohr fringe orientation and an optical wedge’s array arrangement direction, the accuracy error of single-factor angle measurement is sure to be less than 0.01 when the fringe rotation angle is less than 0.1°. Therefore, the above two main errors should be suppressed in the production and assembly so that the measurement accuracy of the interferometer sensor is closer to the high-precision theoretical value.

The Taiji program is a key task for China's space gravitational wave detection, and as an important part of space gravitational wave detection, the telescope's performance will directly affect the accuracy of gravitational wave detection. The existing typical space gravitational wave telescope structure has high secondary mirror sensitivity, which is difficult to meet the manufacturing and adjustment tolerance requirements of larger aperture space gravitational wave telescopes, especially the tolerance requirements for in-orbit stability. In order to solve the above problems, firstly, a new optical system structure of space gravitational wave telescope with intermediate image plane set between three and four mirrors is proposed to reduce the sensitivity of the secondary mirror. Combined with the theoretical method of Gaussian optics, the initial parameters of the structure of the new telescope are theoretically analyzed and calculated. Secondly, through the optimization design, a telescope optical system with a pupil diameter of 400mm, a magnification of 80 times, a scientific field of view of ± 8μrad, and a wavefront error RMS value of better than 0.0063λ was obtained. Finally, the sensitivity evaluation tolerance allocation table of the telescope system is established, and the tolerance of the existing telescope structure and the new telescope structure are compared and analyzed. Compared with the existing telescope structure, the sensitivity of the new telescope structure is reduced by 30.4%. The results show that the new telescope structure has the advantage of low sensitivity, which provides an optimal scheme for the design of space gravitational wave telescopes.

The polarization-multiplexing of a laser based on a medium with a large gain bandwidth and a high thermal conductivity can benefit dual-frequency and dual-comb lasers’ spectral bandwidth and power. This paper presents a demonstration of the polarization-multiplexing of a laser based on a bulk Yb:CALGO crystal. The polarization-multiplexing is realized by sandwiching the gain crystal within two 45˚-cut birefringent crystals, forming two orthogonally polarized beams with spatial separation inside and allowing the system to pump the laser at the waist of the cavity mode for more efficient pumping. The laser outputs watt-level power with a slope efficiency exceeding 30%. A dual-frequency operation with terahertz frequency separation is realized by inserting an etalon into the cavity.

The Taiji Project is a space gravitational wave detection mission proposed by the Chinese Academy of Sciences, which uses the method of laser differential interference to detect pm-level displacement fluctuations caused by gravitational waves between satellites. In order to eliminate the phase measurement error caused by the dissynchronization of the clock between satellites, Taiji Project intends to use the sideband multiplication transfer scheme to measure and eliminate the inter-satellite clock noise. This paper discusses the requirements, principles, and methods of inter-satellite clock noise transmission of the Taiji project, and designs experiments for principle verification. By building an electronics experiment, the limit value of the clock noise of the two systems was tested, the relevant parameters of the experiment were determined, and the principle of the sideband multiplication transfer scheme was verified by further optical experiments. The experimental results show that the clock noise cancellation scheme and related parameters proposed in this paper are reasonable and feasible, and are suitable for the needs of Taiji project. Moreover, in the 0.05 Hz-1 Hz frequency band, the suppression effect of inter-satellite clock noise is better than 2π×10^-5 rad/Hz^1/2, which meets the noise requirements of Taiji pathfinder and lays an experimental and theoretical foundation for the design of clock noise transmission scheme and parameter of future Taiji Project.

A novel photonic quasi-crystal fiber (PQF) sensor based on surface plasmon resonance (SPR) is designed for simultaneous detection of methane and hydrogen. In the sensor, Pd-WO₃ and cryptophane E doped polysiloxane films deposited on silver films are the hydrogen and methane sensing materials, respectively. The PQF-SPR sensor is analyzed numerically by the full-vector finite element method and excellent sensing performance is demonstrated. The maximum and average hydrogen sensitivities are 0.8 nm/% and 0.65 nm/% in the concentration range of 0% to 3.5% and the maximum and average methane sensitivities are 10 nm/% and 8.81 nm/% in the range between 0% and 3.5%. The sensor provides the capability of detecting multiple gases and has large potential in device miniaturization and remote monitoring.