Coherent Doppler lidar (CDL) is frequently used to detect wake vortexes in the clear days because of its advantages of high precision, high spatial and temporal resolution and so on. To improve the observation efficiency of wake vortexes and realize the fast processing and real-time displaying of data, the wake vortex observation experiments were performed in Sichuan province in 2018 and a fast method for identification of wake vortexes was presented based on the radial velocity and spectral width given by CDL. In addition, the wake vortex evolution processes of typical aircraft types were analyzed based on the proposed method. The results show that the wake vortex evolution tendency is similar for all different aircraft types, which first occurs at about 50 m above the ground, then diffuses downward and outward, and the circulation of wake vortexes gradually decreases. However, it is accompanied by the short-time increase of circulation under the influence of the near-ground effect. With the decrease of aircrafts’ weight and wingspan, the dissipation time of wake vortexes decreases, the initial distance between the cores of the right and left wake vortexes gets short, the fusion phenomenon of spectral width becomes more obvious, and the time for the disappearance of this phenomenon increases.

According to the amplification theory of a few-mode erbium-doped fiber amplifier (FM-EDFA), the study on gain simulation is carried out for the four-mode multiplexed signal and the five-mode multiplexed signal. The genetic algorithm is used to optimize the three-layer erbium-doped four-mode group few-mode fiber amplifier and four-layer erbium-doped five-mode group few-mode fiber amplifier. The results of simulation and optimization show that the 1550 nm four-mode multiplexed signal is amplified by using the 980 nm two-mode multiplexing pump and the core pumped forward pump mode. The average gain of each mode is 24.48 dB, and the gain difference between the modes is 0.103 dB. The 980 nm three-mode multiplexing pump is used for amplifying signals, and the average gain of each mode is 23.31 dB. The gain difference between modes is 0.016 dB. By optimizing the pump mode combination and fiber doping structure, the gain performance of four-mode group and five-mode group few-mode fiber amplifiers in C-band is improved.

We proposed a temperature sensor of tapered multimode fiber based on surface graphene modification (SG-TMMF) in this paper. Specifically, a section of TMMF is fused between two sections of single-mode fibers (SMF), and then its surface is coated with a graphene film by the liquid phase transfer method. Graphene interacts with the optical field of the TMMF. The effective refractive index of the SG-TMMF composite waveguide changes with the external temperature. As a result, its optical transmission loss varies, achieving temperature sensing. The experimental results show that graphene can effectively improve the temperature sensing capability of TMMF. In the range of 20-90 ℃, the SG-TMMF with a waist diameter of 9.95 μm has a loss sensitivity up to 0.1589 dB/℃ and a linearity of 0.984. Besides, its sensor has good reversibility. The proposed sensor demonstrating simple manufacturing and high sensitivity has a good application prospect in scientific research and temperature measurement areas in industrial and agricultural production.

Rotary interferometers enable the miniaturization of portable Fourier infrared spectrometers and provide high-speed spectral output. However, their optical path difference (OPD) is nonlinear. In this paper, we analyzed the structure of a portable Fourier-transform infrared spectroscopy (FTIR) rotary interferometer in detail and built a reference coordinate system. Then, an OPD equation for the rotating mirror was deduced according to the geometrical optics principle. Numerical simulation analysis was carried out to illustrate the relationship of OPD with the thickness, refractive index, and rotation angle of the rotating mirror. In terms of sampling without reference laser, the nonlinear relation between OPD and rotation angle was fitted by polynomials to obtain the time non-uniformity corresponding to equal OPD. As a result, the sampling with equal OPD in the case of non-equal time was achieved. Compared with that of the equal-time sampling, the instability of OPD velocity was reduced to 1/5.3, which offered a greater instability margin for the speed control of the rotating mirror and also improved the vibration resistance of FTIR rotary interferometers. Our work provides data support and theoretical basis for the design parameter determination and optimization of portable high-speed FTIR rotary interferometers without reference laser.

The internal wave generated by the movement of the underwater body makes the water surface form weak infrared texture signals, which makes it possible to use infrared means for detection. However, the contrast of texture signals is very low, and it is mixed with the background clutter with large amplitude, which causes great difficulty in signal extraction. Based on the curvelet transform, the curvelet scale component and direction component are screened according to the contrast and frequency characteristics of weak textures, and a clearer texture extraction image is obtained by combining with the threshold optimization and the edge gradient operator. Compared with the results of the traditional curvelet transform, the information entropy and frequency concentration of the image are improved by 30% and 11%, respectively. When the contrast of the weak texture is greater than 5% and the deviation between the direction of the screening frequency and the direction of the weak texture frequency is less than 12°, the algorithm can clearly extract texture information.

Aiming at solving the problems of low segmentation accuracy in the automatic segmentation of brain white matter hyperintensity region on magnetic resonance imaging brain images and easy to miss small lesions, a U-Net segmentation model combining attention and inception is proposed. In the coding stage of U-NET, the Inception module is added to increase the width of the network, so that it has the ability to extract multi-scale features, and the attention module is added to enhance the attention of the network to the segmentation target. The addition and fusion of the two can effectively improve the feature extraction and expression capabilities of the network. Simultaneously, adding residual connections on each convolutional layer in the decoding stage can improve the optimization speed of the network. In addition, because of the problem that sample imbalance easily leads to too many false negatives in the segmentation results, the Tversky loss function with balance adjustment ability is employed to optimize network training. The experimental results show that the proposed method can segment brain white matter hyperintensity region, especially the small lesion area, and each segmentation index is better than those of multiple comparison methods.

By analyzing the on-board lightning data of the FY-4 lightning mapping imager (LMI), there are three types of nonlightning events: ghost noise, high-energy-particle-trajectory noise, and shot noise. To improve the performance of lightning detection, based on an in-depth analysis of the characteristics of false events, the ghost-image-filtering method based on spatial energy correlation, the high-energy-particle-trajectory noise-filtering method based on Hough linear detection, and the shot-noise-filtering method based on spatiotemporal correlation clustering are proposed, and the lightning false-alarm-filtering-processing system method is performed. To test the performance of the false-alarm-filtering algorithm, the lightning-filter algorithm is applied to process two thunderstorms in Yunnan and Hainan provinces. We compare the lightning data obtained by the LMI with those obtained by ground-based network. The results show that the distribution characteristics of lightning events processed by the proposed algorithm are consistent with the ground-based detection results, which verifies the continuous tracking performance of the FY-4 lightning mapping imager for the lightning change process.

In this paper, an approach using a defocused Shack-Hartmann sensor for quantitative phase imaging (QPI) is proposed. In this technology, the defocused Shack-Hartmann wavefront sensor is used to record the intensity distribution under the illumination of light with two different wavelengths, and the dual wavelength phase retrieval algorithm is used for phase retrieval. The digital light field through the phase object is obtained, and the imaging of pure phase objects is realized. Numerical simulation results show that the QPI method is simple and with high accuracy and fast convergence, which is a very promising QPI technology.

The perturbation in a fluid field can be captured by a visualized imaging system. In an optical schlieren system, we theoretically and experimentally compared the images of three-dimensional perturbation in the fluid field. The variations of fluid density generated by three-dimensional ultrasonic field are different in the direction of light propagation. As a result, after the light passes through the density-changing fluid field, the images cannot truly reflect the perturbation in the acoustic field due to the phase accumulation. Focusing schlieren systems with parallel line grids also cannot image the three-dimensional ultrasonic perturbation in the fluid field because of their multiple cutoffs. For this reason, we proposed a modified focusing schlieren system with annulus source and cutoff grids to obtain the information about three-dimensional perturbation. The images obtained by the optimized system achieve the same characteristics with the ultrasonic perturbation in the actual fluid field. In combination with image reconstruction techniques, this system supports the reconstruction of complex three-dimensional perturbation in the fluid field.

This paper proposes to apply a semi-implicit difference scheme where diffusion terms in the time-dependent partial differential equation are discretized implicitly and non-diffusion ones explicitly thus overcoming the prohibitive step-size in lithographic mask optimization (MO). Further, monitoring pixels on mask pattern contour instead of all pattern pixels are selected locally corresponding to high-frequency layout component and optimized to ease the computation complexity. Superior MO performance is demonstrated by the simulation results in terms of improved convergence with reduced optimization dimensionality.

The potential radiation damage in CT scans have been receiving increasing attention. However, reducing the scan dose will degrade the image quality and affect the diagnosis results. Aiming at addressing the above problems, a three-dimensional (3D) reconstruction algorithm combining convolutional sparse coding and gradient L0-norm is proposed herein. The proposed algorithm uses the frequency decomposition reconstruction form to perform unsupervised multiscale online convolution sparse-coding constraints on high-frequency components, and gradient L0-norm constraints on low-frequency components to achieve the suppression and organization of noise artifacts in low-dose CT imaging keep the details. Moreover, three different scales of 3D filter sets are used in convolutional sparse coding, which can effectively adapt to the feature information at different scales and improve the coding ability. The experimental results of abdominal CT simulation data and real-time scan data show that the proposed algorithm can obtain fewer noise artifacts, high contrast in structural details, and better imaging results in the reconstruction process of 25% conventional dose.

Measuring the three-dimensional (3D) shape of metal parts is essential for evaluating their processing quality. The 3D reconstruction scheme based on structured light is affected by the bright reflections from the metal parts, which lead to reduced signal-to-noise ratio and loss of data. To solve the highlighting problem in the detection of metal parts, an adaptive fringe projection measurement scheme is proposed herein and is applied to the measurement of shape and dimension of metal surface. The highlight area of the image is detected by a combined thresholding method and color-space transformation method. The transformation relationship between the camera and projector is obtained by a projection-camera dual-view search method based on an epipolar constraint. The modulation parameters are generated by a scheme based on the bidirectional reflectance distribution function (BRDF) optical imaging model. The BRDF-based scheme realizes pixel-level modulation of the projection fringe. In addition to reducing the number of projections and improving the measurement efficiency, the compensation rate of invalid area obtained by the proposed method is 51.3% in the highlight area measurement results, and the measurement accuracy is improved up 77.4%.

As the digital image correlation technique is gradually applied to the tasks such as long-distance multi-point deflection monitoring of engineering structures, camera shake in the non-controlled measurement scenarios such as outdoor scene has become one of the main reasons that affect the measurement accuracy. In order to eliminate the influence of camera shake on target displacement measurement, this study firstly establishes an affine transformation model that represents the camera disturbance movement at a long distance based on the projection model of the pinhole camera movement, and then combines the stable reference points to correct the image point coordinates after the camera movement. To verify the effectiveness of the affine model, three sets of translation experiments are carried out. The experimental results show that when the working distance is greater than 2 m, the affine transformation model effectively eliminates the measurement errors caused by camera shake in a simple way and is not influenced by the depth distributions of the reference and target points.

In this paper, we propose an evaluation index for optical circulator noise and explained the principle of laser heterodyne interference. Then, we establish a noise transfer model for a typical circulator in the displacement measurement by laser interferometry systems. In addition, in combination with the noise source of the circulator, we demonstrate the influence of the nonlinear error coefficient of the circulator and the phase of stray light on the displacement measurement error. Furthermore, we propose an error evaluation method with the circulator directivity as the noise evaluation index and verified its applicability by numerical simulation. Aiming at the main source of noise of the circulator, a kind of point diffraction spatial optical circulator is designed and developed based on theoretical analysis model. This circulator has a directivity of 129 dB, the directivity of circulator is 66 dB higher than that of typical commercial circulator, thus achieving high-precision measurement of target displacement.

Although the traditional Fredkin gate can achieve the corresponding logic function, its extinction ratio and crosstalk need further improvement. In view of this, a silicon-based all-optical Fredkin gate based on the cross-phase modulation effect is proposed and designed. The all-optical reversible logic gate consists of two 2×2 directional couplers, a 2×1 directional coupler, and a 1×2 directional coupler and two-phase shift arms. By using the cross-phase modulation effect caused by the pump light and the signal light in the phase shifter arm, the phase difference of the signal light in the upper and lower phase shift arms can be changed, outputting light waves of different amplitudes at different ports of the designed device. Then, it realizes the logic function of the Fredkin gate. Besides, MATLAB and integrating the split-step Fourier method are used to simulate and analyze the designed silicon-based all-optical Fredkin gate. The simulation results show that the worst extinction ratio of the device can reach 48.46 dB.

This paper theoretically studies the design of a device for achieving asymmetric transmission of circularly polarized light. The device is a two-dimensional (2D) photonic crystal (PhC) heterostructure with a fully photonic bandgap, which is with triangular lattice air holes embedded in the germanium and silicon. Here, a line defect is introduced in the 2D PhC to form an optical waveguide structure that can achieve high forward transmittance. At the same time, a microcavity structure is designed to diverge the light, which is combined with the total reflection principle to suppress backward incident light, and therefore the asymmetric transmission of circularly polarized light is achieved. As a result, the asymmetric transmission of circularly polarized light at the telecommunication wavelength (1550 nm) with high forward transmittance (up to 0.726) is realized. As the circularly polarized light is a linear superposition of two orthogonally linear polarization lights with a fixed phase difference (π/2), the structure designed in this study can realize asymmetric transmission of arbitrarily linearly polarized light at the same time. Therefore, it has a wide range of applications, including quantum communication, information processing, and integrated optics.

Based on the theory of q-parameter, the collimating factor of the Gaussian beam transformed by the inverted telescope system has been strictly deduced. Assuming the aperture of lens is larger than beam size, the collimation factor of the system is determined by the geometric compression only, instead of waist radius or object distance. Further research found that under the appropriate approximation premise, the derivation results in this paper can be unified with the relevant derivations in the domestic laser principle textbook. The proposed algorithm is simulated, and the results show that the collimation factor obtained in this paper is consistent with the results obtained under the conditions discussed in the laser principle textbook. The collimation factor of the Gaussian beam through the inverted telephoto system obtained in this paper can provide a more accurate reference for the collimation operation part of related scientific research and engineering practice activities.

The existing monocular-based three-point laser vision measurement system is susceptible to errors such as laser feature point 2D image detection, camera distortion, image center point coordinate, and focal length errors. Aiming at alleviating these errors, a new ranging and pose estimation is proposed. In the proposed method, the solution of the attitude angle is only related to the distance value, fixed parameters of the camera, and installation position of three lasers. First, the three lasers are controlled to calibrate the internal parameters of the monocular camera when emitting light in parallel, and the pose coordinates of the center point of the camera are calculated. Then, according to the calibration parameters of the camera at different distances, an iterative method is used to solve the minimum error. Finally, the downward tilt γ angle of the top laser is adjusted to calculate the yaw and pitch angles of the target plane. The experimental results show that the accuracy of distance measurement and attitude estimation is ensured while reducing the influence of the error term. The average error of distance measurement within a distance of 200-600 mm is 0.65 mm, and the error of yaw angle and pitch angle is less than 0.98°.

Aerospace products have large structural sizes, complex surface shapes, and high assembly accuracy requirements. As automated assembly equipment, mobile robots can achieve precise positioning of aircraft surface crawling and moving assembly poses as well as compensate the relative positioning error between the tool center point of the hole-making robot and the workpiece. Aiming at the requirement of the three-dimensional recognition and measurement positioning of assembly datum under the core clip interference, we propose a high-precision positioning method for the hole-making datum of the orbital crawling robot. First, the adaptive contour extraction algorithm based on gray-scale clustering is used to achieve the segmentation, recognition, and coordinate calculation of the reference hole contour. Then, the three-dimensional conversion of the reference hole coordinates is realized using the target-type high-precision hand-eye calibration method. Finally, a visual measurement system is integrated on the hole-making equipment, and field testing and accuracy verification are carried out. Experimental results show that the positioning error of the reference hole of the method is less than 0.05 mm, thus meeting the assembly and positioning requirements of the orbital crawling hole-making robot.

With regard to the low spatial resolution in light filed microscopy, we proposed a method based on graph regularization and reconstructed a super-resolution light field. As a result, high-resolution view angle images of the light field were obtained. Firstly, the images of the light field were aperiodically extracted. Then, the super-resolution issue in the light field was transformed into a global optimization problem, in which the complementary information between those images was applied for smoothing regularization. Finally, the gradient descent algorithm minimized the objective function and thus reconstructed the high-resolution view angle images. Experimentally, a light field microscope, equipped with an objective lens, a microlens array, and a CCD, collected the data of the light field, and then graph regularization was used to reconstruct the high-resolution light field. Compared with the traditional methods, the proposed method has small calculation amount in disparity estimation and high imaging quality, and effectively retains the original light field structure.

Based on the vector wavefront aberration theory, the field curvature characteristic of the off-axis three-mirror optical system under a small misalignment during installation and adjustment was analyzed. A method for generating field curvature by tilting the focal plane to compensate for system misalignment is proposed. The simulation analysis was conducted using a remote sensing camera. Besides, the relationship between the number of split field-of-view of focal plane and the tertiary mirror installation accuracy requirements was analyzed, and the optical system alignment process and method were given. Then, the field curvature characteristic of the camera was tested by measuring the modulation transfer function, and the focal plane inclination angle was corrected to compensate for the curvature aberration. After the correction, the modulation transfer function of the edge field-of-view significantly improved. The modulation transfer function of each field-of-view of the camera is greater than 0.21.

A fixed free-form lens array corrector is proposed in this paper to correct the dynamic aberration introduced by a conformal dome with large length-diameter ratio and large scanning angle. By designing the free-form surface shape and rotation angle of each lens of the lens array corrector, the dynamic aberration introduced by the large length-diameter ratio conformal fairing in different scanning angles is corrected. The experimental results show that the free-form lens array corrector can realize ±50° scanning angle dynamic aberration correction for a conformal dome with a length-diameter ratio of 1.2. The new corrector has the advantages of both a dynamic corrector and a fixed corrector, and is suitable for a conformal dome optical system with a large scanning angle, a large length-diameter ratio and high stability.

The optical interconnection speed of data center is developing rapidly in recent years. As for the optical interconnection of data center, a compact, low-loss, and small output mode field four-channel coarse wavelength division demultiplexer chip is designed and fabricated, which is using quartz-based silica optical waveguide with 1.5% refractive index difference. This chip can meet the transmission velocity requirement of 200 Gbit·s -1/400 Gbit·s -1 for high-speed data center. The minimum insertion loss is less than 1.07 dB, the 1 dB bandwidth is larger than 13.7 nm and the 3 dB bandwidth is larger than 16.1 nm. The polarization-dependent loss is smaller than 0.08 dB, the adjacent crosstalk is larger than 24 dB, and the non-adjacent crosstalk is larger than 30 dB. The designed chip can fully meet the commercial requirements of a wavelength division demultiplexer chip for optical interconnect of high-speed data center.

For the difficulty in measuring the distribution characteristics of PM2.5 concentration in the atmosphere, we used 532 nm lidar to continuously observe the Huainan area from June 1st to December 31st, 2016. A regression prediction model was established concerning the atmospheric boundary layer height, aerosol optical depth, temperature, relative humidity, wind speed, visibility, and measured PM2.5 concentration to identify the PM2.5 concentration. Since the traditional backpropagation neural network (BP) was prone to the local minimum, we adopted a genetic algorithm-based backpropagation neural network (GA-BP) according to the data characteristics and applied the genetic algorithm to finding the optimal weights and thresholds, balancing global and local contradictions. A comparison of the two regression models demonstrates that the GA-BP method is significantly better than the BP method. The correlation index R2of the test set and the mean forecast error are respectively 0.623 and 24.692 μg/m 3 for the BP method, and 0.899 and 7.122 μg/m 3 for the GA-BP method. These results indicate that lidar can effectively monitor the PM2.5 distribution in the atmosphere and provide data support and reference for the monitoring of atmospheric PM2.5 in the Huainan area.

The scanning mirror is an important component of hyperspectral imaging LiDAR. Three scanning modes of polyhedral rotating mirror scanning, vibrator mirror scanning, and conical scanning are analyzed in this paper. Traces of different scanning modes are deduced, and the relationship between laser scanning traces and scanning modes is obtained by combining aircraft tracks. On this basis, a conical scanning mode for hyperspectral imaging LiDAR system is designed. Experimental results show that when the flight altitude is 500 m, the repetition frequency of the laser pulse is 130 kHz and the divergence angle of the laser pulse is 0.3 mrad. The method meets the requirements of the field angle of 30° and the angle between two adjacent pulses is less than 1 mrad. Compared with other scanning methods, conical scanning can improve the density of LiDAR spot footpoints, which provides help for effective data collection and accurate classification of ground objects.

The visible-shortwave infrared advanced hyperspectral imager (AHSI) loaded on the GF-5 satellite can acquire information about 330 spectral bands, which facilitates the derivation of land surface properties by deploying the hyperspectral observations of both atmosphere and land surface. However, cloud contamination in remotely sensed images often limits its application. To improve the availability of the GF-5 data, this study proposed a cloud detection approach that can be applied to various situations using the hyperspectral data proved by GF-5 AHSI. The apparent reflectance at the top of the atmosphere was firstly derived with the Level-1 product by the associated radiometric calibration coefficients and spectral response function. We found that the thick cloud pixels in the images can be effectively distinguished from other land cover types at the visible spectral region after the comparison of their apparent reflectance. The broadband apparent reflectance derived with the corresponding narrow bands was used to detect thick cloud pixels, which can eliminate the impact of noise associated with the narrow-band data. On this basis, the thick clouds can be screened out using simple detection thresholds. We then obtained the candidate thin cloud pixels using the cirrus cloud band. As thin cloud pixels were generally confused with high-albedo pixels, the distinction between these two features was studied by comparing the band ratios in various combinations. The thin cloud pixels were finally detected based on the optimal band combination and the corresponding threshold. Furthermore, we adopted the visual interpretation of cloud pixels to evaluate the performance of our algorithm using several GF-5 AHSI images. The cloud pixels can be well distinguished from clear sky pixels with an accuracy of over 91%, which indicates that our approach can be used to accurately detecting clouds for hyperspectral remote sensing images.

According to Locard's switching theory, fiber evidence can help reconstruct criminal incidents. Generally, fiber evidence is identified by chemical composition analysis; however, the integrity of material evidence will be destroyed. For this reason, we established a testing and analysis method based on the theory of bidirectional reflectance distribution function (BRDF) to ensure the long-term retention of fiber evidence. To be specific, three groups of fiber samples were selected, with each group containing six kinds of fiber samples with the same color and different materials. The first-order differential and difference methods were used to extract spectral characteristics and analyze spectral BRDF. According to spectral curve fitting, the fitting function and residual coefficient were extracted for differential comparison. From the physical mechanism, a Davis model was built to simulate the spectral BRDF curve. As a result, the material evidence and suspicious results in the cases were effectively determined. The experimental results provide an effective and non-destructive detection technology and analysis method for the identification of fiber evidence characteristics, which greatly facilitates criminal investigation analysis and conviction.

At present, the optical losses of crystalline silicon heterojunction solar cells are mostly reduced with textured surfaces. However, this method has a complicated process and poor repeatability and film uniformity; meanwhile, the textured surfaces increase the carrier transfer path and electron-hole recombination, which hinders the improvement of cell performance. In this paper, a double-layer TiO2/SiNx antireflection film on planar silicon substrate was designed by OPAL and TFCalc. Considering the solar spectrum distribution and the spectral response of the crystalline silicon heterojunction solar cells, we optimized this antireflection film by taking the film, glass, and substrate as a whole, with the weighted average optical loss as the evaluation function. Furthermore, the designed film was compared with a single-layer ITO antireflection film on textured silicon in terms of the weighted average optical loss. The results show that the weighted average optical loss of the double-layer antireflection film was 4.69%, which was about 1.97 percentage points lower than that of the single-layer ITO film on textured silicon. Also, the absorption loss of the double-layer film significantly dropped. Our findings provide theoretical support to replace textured silicon with planar silicon.