Herein, the optical path difference (OPD) distribution of supersonic semi-free jet under different total pressure conditions was investigated in a vacuum chamber to address the problem of reduction of image quality attributed to the non-uniform distribution of jet refractive index field. The angle between the main optical axis of the three optical paths and the flow direction was 81°, 90°, and 99°. Experimental results show that larger total pressure corresponds to larger OPD of the jet. Under the same total pressures, the OPD of the 81° optical path is larger than the 99° and 90° optical paths. An empirical polynomial of the OPD along the flow direction was deduced, and the applicability of two theoretical models in the jet flow field was also analyzed.

A wavelength detection method based on gated recurrent unit (GRU) network was proposed to demodulate the central wavelength of overlapping spectra in fiber Bragg grating (FBG) sensor networks. The proposed method transformed the wavelength demodulation problem of overlapping spectra in FBG into a regression problem and considered the sequence and spectrum characteristics of the spectral data. To learn the spectral data characteristics and train to achieve the corresponding wavelength detection model, the GRU network was used. Thus, the overlapping spectra could be quickly and accurately demodulated. Experimental results show that the proposed method can overcome the precise demodulation problem of the central wavelength for partially or completely overlapping spectra of FBG sensor networks. The test results with root mean square less than 1 pm account for 88.2% of the test results. The detection accuracy and stability of the proposed method provide enhanced results compared with existing demodulation methods. The proposed method provides a novel way to improve the multiplexing capability of FBG sensor networks.

Random polarization control in Golay-coded Brillouin optical time domain analysis (BOTDA) fiber sensors can be used to mitigate the deterioration in the signal-to-noise ratio caused by the strong pulling effect of polarization. However, a relatively strong polarization random noise (PRN) is induced, which may reduce the measurement accuracy of the sensing signal. In this study, the effect of PRN on the sensing signal was confirmed through simulation analysis and experimental verification. Furthermore, Brillouin phase shift was used to suppress PRN. Experimental results show that the sensing accuracy is enhanced threefold using the proposed method.

In this study, we investigate the visible light communication technology under ultralow illumination to satisfy the wide coverage and multiconnection requirements associated with the internet of things. A light-emitting diode (LED) is used to transmit the visible light signals in the infrared protocol format. The calculations denote that the human eye can not notice flicker when its modulation depth is less than 0.625%. Further, an indoor visible light communication scene (30 m×2.1 m×2.6 m) is simulated by considering a Lambertian source radiation model; the signal coverage become 616 m 2 when a 9.7 W LED illumination lamp with a 20×12 array is used as the transmitting device and the modulation depth is 0.46%. Subsequently, a visible light-based intelligent home system is developed in the laboratory to estimate the farthest distance at which a remotely controlled robot can respond to the signal with 100% accuracy. The relative error between the experimental results obtained at 14.3 m and the simulation results obtained at 14 m is 2%.

To extensively study the applicability of Fiber Bragg grating (FBGs) sensing technology in the penetration test of jacked piles, a large-scale model test system is used to conduct the penetration test of jacked piles. This paper uses a fiber grating strain sensor for data testing, while using a spoke type pressure sensor and a fiber grating pressure sensor as control groups. By installing a full-section dynamic spoke pressure sensor and a fiber grating pressure sensor at the end and top of the pile, respectively, the difference between the penetration processes of the jacked piles based on the two sensing technologies is compared and analyzed. Simultaneously, the pile body strain data is collected based on the FBG sensing technology, and the penetration characteristics of the side frictional resistance, the axial force, and the unit side frictional resistance of the pile are studied. Results show that the FBG sensing technology has superior performance in the penetration test of static pressure piles, can accurately reflect the penetration characteristics of jacked piles, and can clearly reflect the pile jacking-in process with the increase of pile driving length. Furthermore, it is found that the driving pressure, the end resistance, the axial force, and the unit side frictional resistance of the pile body have the characteristics of a steady-state penetration. Therefore, this study presents a great reference for the model test and the engineering design of static pressure piles.

To solve the over-smoothing problem of image details caused by fixed filtering coefficients during the filtering process in the traditional non-local means (NLM) algorithms, a weight function comprising an adaptive filtering coefficient is designed using the structural tensor (ST) trace as a discriminant criterion of image feature areas and called as ST-NLM. Meanwhile, to solve the time consuming problem of the traditional algorithms, the proposed algorithm is accelerated by integral images. The test results demonstrate that the overall smoothness and detail retention of images are relatively good after denoising by the ST-NLM method. Compared with those by the NLM method, the peak signal-to-noise ratio, structural similarity, and running speed by the ST-NLM method increase by 3 dB, 5%, and twice, respectively.

The optical coherence tomography (OCT) technology has been extensively used in clinical medicine, (including ophthalmology, intravascular endoscopy) and pharmacology, because of its non-invasive and non-contact features. Until now, OCT can be categorized into two major types according to the priority of imaging direction. One is the standard OCT which can perform A-line scanning along the incident direction of sample beam and has the ability to generate the longitudinal section (B scan) image, normally used to identify the individual retinal layers. The other is the en-face OCT (also called frontal section OCT) which can generate the transverse section image of sample layer in the direction perpendicular to the incident light. Using en-face OCT, the fine structure of the biological sample can be displayed in a field-of-view similar to that of the microscope, considerably extending the acquisition and representation modes of OCT imaging. Different signal acquisition methods can be adopted by en-face OCT systems. A detailed analysis and summary regarding these methods as well as the developing roadmap of en-face OCT are provided.

In this paper, a passive quasi-optical terahertz imaging system for a human body based on a single detector with fast scan mechanism is presented. A resolution test chart for image quality evaluation is designed. The proposed imaging system comprises several reflection quasi-optical devices. Terahertz waves emitted from the human body were transmitted through a terahertz window into the imager and were sequentially reflected by a flapping mirror, rotating mirror, focusing mirror, and fixed mirror and finally reached the terahertz detector. The imaging system can obtain a full image of 0.65 m×1.9 m using only one detector, with a resolution of 2 cm and an imaging distance of 1.5 m within 2.5 s using a fast scanning mechanism comprising a flapping mirror and a rotating mirror. The imaging system is capable of displaying the profiles of suspected dangerous items carried on the human body, such as metals, liquids, ceramics, and powders. The detected items are automatically marked by software.

In this study, the dark-field detection algorithm, which is suitable for the detection of contaminants, is investigated in accordance with the imaging characteristics of the surface contaminants of the large-aperture reflector. In this algorithm, the autofocus algorithm is considered during the image acquisition process, whereas the distortion correction and pollutant extraction algorithms are considered during the image processing process. Further, the Tenengrad function is selected to evaluate the sharpness during the autofocus process, and a coarse-precision peak search strategy is proposed to improve the focusing accuracy. Based on the distortion model, the distortion correction algorithm calculates the distortion model coefficients in accordance with the projective transformation properties of the calibration plate corner points and implements image distortion correction. The root mean square error of the correction result is 3.3092 pixel. In the contaminant extraction algorithm, the top-hat transform is employed to eliminate the image background, and the Laplacian weighted adaptive binarization algorithm is used to extract contaminants from the background-removed image. The algorithm is effective for the image with small-sized pollutants in case of uneven illumination. The error in the amount of detected contaminants is 7%. The detection accuracy of the proposed method is better than those of the global threshold algorithm and the mean operator weighted adaptive binarization algorithm. Furthermore, the detection algorithm can provide technical support to evaluate the clean state of the reflector.

A hyperspectral Mueller matrix imaging (HMMI) method to capture spatial, spectral and Mueller matrix images at the same time is proposed. The principle of hyperspectral Mueller matrix imaging and the shear interference imaging process of birefringent interferometer are discussed in detail. The joint optimization design of polarization state generator and polarization state analyzer as well as the calibration method of this system is shown. In order to verify the performance of the instrument, the spectral Mueller matrix imaging of the target in the laboratory proves the feasibility of HMMI in quickly acquiring spectral images and Mueller matrix images. Because of its high spectral resolution and fast polarization modulation, it provides a new idea for the development of spectral Mueller matrix imaging.

Aiming at the problem that the 3D information captured by a light field camera can only provide a very narrow parallax when it is displayed, the recording process of information by the light field camera is analyzed, and a full-parallax 3D display method is proposed which converts the original light field data into an element image array (EIA) and resamples the objects in EIA to adjust the depth. In order to verify the validity of this method, the real 3D object is captured by the light field camera and displayed in the depth priority integral imaging (DPII) system after processing the light field image with the proposed method. The experimental results show that the full-parallax 3D display of light field data can be realized by the proposed method.

Based on the principles of particle size measurement by multi-wavelength light extinction and image methods, a method for synchronous measurement of trans-micron scale mixed particle size is proposed, and the extinction spectral inversion algorithm for submicron-ten microns particle size and the image processing algorithm handling particle size over ten microns are established. The device for synchronously measuring extinction spectra and backlight images is established with a dispersion prism. The experiment with the trans-micron scale mixed particle samples prepared by ten kinds of standard particles with the sizes of 500 nm-76.9 μm is conducted. The results show that when the method is used to synchronously measure the trans-micron scale mixed particle size, the interaction between the extinction spectrum of submicron-ten microns scale particles and the backlight image of particles over ten microns can be ignored, and the trans-micron scale mixed particle size can be measured synchronously. The extinction method and the image method are respectively used to measure submicron-ten microns particle size and particle size over ten microns, the relative errors are less than 8% compared with those with the standard particle size, and the measurement repeatability is good, which provides an effective particle size measurement method for trans-micron scale mixed particles.

The scanning Hartmann technology is a commonly used method to detect the imaging quality of large-aperture telescopes; however, the detection performance of different-order aberrations and the detection accuracy under different sub-aperture distributions remain unclear. Therefore, we develop a simulation model using Matlab and Zemax to study detection performance of this technology. The simulation results show that the highest 28 th-order aberration can be detected using the scanning Hartmann technology and that the root mean square (RMS) relative error is less than 5%. Further, it is difficult to distinguish the high-order aberration component when detecting multi-order aberrations. The usage of tangent sub-aperture distribution can better balance the detection accuracy and efficiency. The usage of a large number of sub-apertures can effectively increase the detection accuracy; however, the accuracy is slowly improved after the number of sub-apertures is increased to a certain number, while the detection time is greatly increased.

Directional polarimetric camera (DPC) is used to monitor several parameters related to the global atmospheric aerosol and clouds as it can detect atmospheric multi-angle polarization. However, it is an ultra-wide-angle optical instrument; therefore, the polarization effect of lens cannot be neglected, and it should be corrected during the calibration stage. A polarization measurement performance test is conducted prior to launching the instrument in the laboratory and outfield. To meet the requirement of an on-orbit measurement, an on-orbit detection method is designed to assess the polarization detection performance of the optical polarization remote sensing instrument with a large field of view. The rapid evaluation of polarization detection performance in full field of view is achieved by analyzing the polarization rainbow data of ocean flares and water clouds in a wide range of complex environments, and the data processing effectiveness of ground application system is verified. The on-orbit polarization detection performance of DPC is consistent with the laboratory and outfield data, which provides an important basis for aerosol and cloud inversion applications. Furthermore, the detection methods in each stage can be used as a reference in wide-angle polarization remote sensing applications.

A staggered metasurface structure that consists of double-periodic supercells with eight different V-shaped antennas is proposed and its properties are studied theoretically and experimentally. The calculation shows that co-polarized transmission and cross-polarized anomalous refraction can be obtained when the transverse electric light is incident vertically. The phase difference of anomalous refraction can be changed and its amplitude can be regulated by adjusting the transverse misalignment distance between two periodic supercell units. The experimental results show that, when the 4.3 THz light is vertically incident on the samples with two supercells staggered by 0, 2, 4 and 6 V-shaped antenna unit cells, the anomalous refraction intensities are 3.6%, 1.7%, 0.7%, and 1.9% of the incident light intensity, respectively, which is consistent with the calculated results.

An all-optical passive synchronous laser with master-slave configuration is built, the output power of the master laser pulse is amplified and injected into the slave laser, and the pulse synchronization between 1029.9 nm pump light and 1585.5 nm signal light is realized by using the cross-phase modulation effect of the injected pulse in the slave laser. An acousto-optic modulator used for frequency selection together with a cascade fiber amplifier enhances the peak power of pump pulse. In addition, the optimization of fiber link length effectively controls the spectral width broadening effect of pump pulse. The two-color synchronized pulse is processed via nonlinear difference frequency in PPLN crystal. When the repetition rate is set as 100 kHz, a linearly polarized picosecond pulse is achieved with 3-dB spectral width of 0.77 nm, central wavelength of 2940 nm, the maximum single pulse energy of 1.8 μJ , and pump light conversion efficiency of 49.6%.

To meet the illumination system requirements of 45 nm and below node lithography technology, the micromirror array (MMA) used in the beam shaping unit of deep ultraviolet lithography illumination system is used as the key device to produce the freeform source required by the source-mask optimization(SMO) technology. Based on the structural parameters of MMA as well as the manufacture and adjustment characteristics, the angle error types of MMA are analyzed. On this basis, the Monte-Carlo tolerance analysis method is used to simulate the actual manufacture and adjustment processes. After the influence of the micromirror angle error on the exposure results is investigated, the angle tolerance that meets the exposure requirements is established. The results show that when the angle adjustment tolerance and the process angle tolerance of MMA in the orthogonal direction are within the scope of (±0.04°, ±0.06°) and (±0.04°, ±0.04°), respectively, the critical dimension error (CDE) obtained by exposure is less than 0.33 nm at a confidence probability of 98.1%.

In this study, we design an optical system for a visible-infrared (VNIR-SWIR) Offner imaging spectrometer for the space exploration of asteroids. The operation waveband of the proposed system covers from 0.4 μm to 2.7 μm, and the F number is 6 and 3 for the visible/near-infrared (VNIR) band (0.4--1.0 μm) and the short wave infrared (SWIR) band (1.0--2.7 μm), respectively. Subsequent to the initial structure development and optimization, spot diameters in the spot diagrams of the proposed optical system are observed to be less than one pixel, and the minimum modulation transfer function (MTF) is 0.51 at the Nyquist frequency. To improve the spectral response to fulfill the broadband spectral imaging requirement, the convex grating is divided into two regions with different grating periods. The inner region works in the VNIR band, while the other works in the SWIR band. The prototype of the imaging spectrometer optical system is developed and its spectral performance is characterized. Results show that the performance of the proposed prototype is satisfactory and can meet the requirements.

Particle image velocimetry (PIV), as a non-contact, global indirect hydrodynamics measurement technique, can capture the velocity field of a fluid from an image to reveal the motion of the fluid. The development of deep learning technology and its use for PIV have significant research value and a potentially wide range of applications. In this paper, the authors propose an improved lightweight convolutional neural network based on the optical flow neural network. The proposed method improves the accuracy of particle image velocity measurement while reducing the parameter quantity of the model and improving the test speed. First, this work improves the optical flow neural network architecture with superior rigid body estimation performance, and uses an artificial particle image dataset for supervised training. The trained network model is then compared with a state-of-the-art PIV deep learning model. Experimental results indicate that the PIV based on the lightweight deep learning model proposed in this paper can reduce the number of model parameters by 9.5% and improve the test speed by 8.9% without losing accuracy.

In order to study the effect of fog on quantum communication in free space, the effects of fog visibility and transmission distance on link attenuation, quantum communication channel capacity, channel fidelity and channel error rate are studied according to the distribution function and extinction coefficient of fog drop spectra. The simulation experiment is also conducted.The results show that when the transmission distance is 9 km and the visibility is 0.2 km and 0.6 km, the corresponding link decay, channel capacity, channel fidelity, channel error rate are 0.76 and 0.25, 0.24 and 0.4, 0.91 and 0.97, 0.01 and 3.26×10 -7, respectively. It can be seen that when the visibility of foggy days is less than 1 km, the influence of fog on the performance is more significant. Therefore, the parameters of the communication system should be adjusted to ensure the smooth progress of communication based on fog visibility.

In order to analyze the measurement-device-independent quantum key distribution protocol more comprehensively, the statistical fluctuation analysis of measurement-device-independent quantum key distribution protocols based on heralded pair coherent state is carried out. First, with the increase of the number of transmitted signal pulses in the statistical fluctuation of light source, the relationships of bit error rate and key generation rate with transmission distance are analyzed. The results show that increasing the number of pulses can improve the key generation rate and the maximum transmission distance, and can reduce the bit error rate. Moreover, the measurement-device-independent quantum key distribution protocol based on heralded pair coherent state has better performance than the one based on heralded single photon sources. When the light source is statistical fluctuating, the relationship between the key generation rate and the transmission distance of measurement-device-independent quantum key distribution protocol based on heralded pair coherent state in asymmetric channels is further analyzed, and the simulation results show that this protocol in asymmetric channels has better performance than that in symmetric channels.

The millimeter accuracy of a satellite laser ranging (SLR) system is often hindered by the satellite signature effects. Based on the actual distribution of the satellite retroreflectors, the retroreflector uneven distribution (RUD) model is proposed in this study. The relationships between laser incident angles and the number, distribution, and reflection intensity of the retroreflectors that have participated in the SLR reflection are calibrated and discussed, while the effects of satellite signature on the echo signal residual and center-of-mass correction (CoM) are analyzed in detail. Results indicate that compared with the common CoM model, the participant retroreflectors are dispersedly and asymmetrically distributed, and their number is a function of laser incident angles. The reflection intensity for each reflector at the same satellite latitude is different and the retroreflected pulse is observed to be broader than the transmitted pulse with significant tailing effect. The above results agree with the experimental results. In addition, the CoM value obtained using RUD model dynamically varies with the laser incident angle. For Lageos-1, the average CoM of Changchun Station is 248.1 mm, which is consistent with the reference range accepted by the international laser ranging system.

The simulation experiments of non-coherent detection of Doppler frequency shift and aerosol are carried out with the Mach-Zehnder interferometer as the spectral analyzer. Under the condition that the pulse energy of the transmitter reaches 500 mJ, the pulse repetition rate is 10 Hz, the spectral line width does not exceed 0.005 cm -1, the beam divergence angle is less than 0.10 mrad, the lens diameter of receiving telescope is 350 mm, the avalanche diode is used as the detector, the digital sampling rate is 2×10 7 sampling·s -1, and the number of sampling data bits is 16 bit (the effective number is 11 bit), the wind profile with the detection distance of 2500 m in 45° oblique range is obtained and the detection accuracy of line-of-sight wind speed is 2 m·s -1. The simulation results show that the radar with the Mach-Zehnder interferometer as the spectral analyzer can be used for the direct detection of Martian wind-sand. Compared with the 354.7 nm laser pulse, the 1064 nm laser pulse (compatible with the laser-induced breakdown spectrometer) has a high interference contrast, and the Mie backscatter strength is higher than the Rayleigh backscatter strength.

The telemetry LIBS system including a coaxial Schwarzschild telescope is built to study the changes of the LIBS characteristic spectral line intensity, the relative standard deviation, the spectral similarity, and the plasma temperature, when the sample distance fluctuates at different guide positions, and to analyze the reasons for the changes of characteristic parameters based on the physical mechanism. The results show that the sample position fluctuation has a significant influence on the plasma temperature, the characteristic spectral line intensity, and the relative standard deviation, while the spectral similarity remains stable in a certain range. Under the condition that the spectrum satisfies a certain similarity, the tolerance of the sample position fluctuation increases linearly with sample distance. When the current system focus range is 1.9-4.1 m and the spectral similarity is 0.99, the range of the sample position fluctuation tolerance is 70-220 mm. The results are beneficial to the design of high-performance optical systems and provide theoretical references for qualitative and quantitative analysis of spectra.

To address the problem of gas infrared spectral identification, a new lifting algorithm named eXtreme gradient boosting (XGBoost) is introduced. Infrared spectral data of chloroform, p-xylene, and tetrachloroethylene are selected for experiments. After these original data are preprocessed, the spectral features are first extracted by feature engineering to generate feature vectors. Then, the XGBoost model is established and its parameters are optimized. Finally, based on a classification accuracy index, the XGBoost model is compared with random forest (RF), support vector machine (SVM), feedforward neural network (FNN), and convolutional neural network (CNN). The experimental results show that XGBoost has a broad application prospect in the field of gas infrared spectral identification.

Based on the non-resonant photoacoustic spectroscopy and the non-porous membrane oil-gas separation theory, we design a miniature detection module by integrating a photoacoustic gas sensor with an oil-gas separation membrane to satisfy the demand for online monitoring of the oil-immersed power transformer. This design can be realized by combining a polytetrafluoroethylene propylene (FEP) polymer membrane with a diameter of 40 mm with a photoacoustic chamber with a volume of only 0.3 mL. The designed module exhibits a small size, short oil-gas separation time, and real-time online detection. When the fault characteristic gas dissolved in the oil diffuses into the gas chamber, the techniques such as near-infrared laser photoacoustic spectroscopy, wavelength modulation, and second-harmonic detection are used to achieve high-sensitivity gas detection. The experimental detection results of the characteristic C2H2 gas prove that an oil-gas separation balance in the oil sensing system with dissolved gas can be realized within 3 h at an oil temperature of 60 ℃. Furthermore, the measurement error is within ±30% for the C2H2 gas dissolved in oil with a volume fraction of 10 -6.