With the fine-mode state M4 and coarse-mode state M9 in the moderate-resolution imaging spectroradiometer (MODIS) ocean aerosol model as examples, the single scattering characteristics of the nonspherical particles with ellipsoidal shape distributions under different aspect ratios and equal-probability aspect ratios are simulated based on the T-matrix scattering theory. Different from the spherical particles, the nonspherical shape causes polarization for fine-mode particles and depolarization for coarse-mode particles. The offshore sea remoting data from the airborne advanced atmospheric multi-angle polarized radiometer (AMPR) are used to retrieve the optical depths of both spherical and nonspherical aerosols. The retrieval results show that the optical depth obtained based on the nonspherical mode is well consistent with the measurement results by the sun-sky radiometer on earth (Cimel,CE318)and the relative root-mean-square error is smaller than 0.1.

The effect of aurora on limb infrared radiation is simulated and analyzed with the infrared radiation model of the middle and upper atmosphere. The simulation results show that when aurora disturbs the atmosphere, the limb radiance at 2.7 μm and 5.3 μm resulted from NO is significantly larger than the quiescent value. In addition, the limb radiation at 2.7 μm is not time-dependent, but that at 5.3 μm has a small time-dependent variance. Moreover, the limb radiance at 4.3 μm resulted from CO2 and NO+ has the obvious increase compared with the quiescent value, and CO2 makes it time-dependent. The radiance at 10 μm and 15 μm is enhanced with a certain amplitude because of the linkage effect from limb radiance at CO2 4.3 μm and similarly the variance is also time-dependent. The chemical reactions introduce the high energy levels of NO and NO+ and thus the radiance at 2.7 μm, 4.3 μm, 5.3 μm gets enhanced at the broad spectral bands including their baseband and hot bands. The properties from the model analysis are applied to the preliminary simulation of an observation event of limb radiation at 4.3 μm when aurora disturbs the atmosphere. The simulation results are compared with the measured data and thus the correctness of the model is verified.

Atom flux is one of the main factors influencing the quality of nano-gratings fabricated by atom lithography. Based on the theoretical model of eruption volume of atomic furnace tube, the atom flux levels for three typical kinds of furnace tubes are compared by the combination of theory and experiment. Moreover, with the best furnace tube configuration, the peak to valley height of the fabricated Cr nano-gratings increases up to 100 nm, which further optimizes the quality of nano-gratings.

To recognize the point target maneuvering mode from the spectral characteristic dimension, the mapping relationship between the point target maneuvering mode and the spectral signals is built, and the multi-spectral radiation characteristics of point target with a maneuvering status in the direction of observation are investigated. The features of multi-spectral radiation signals are extracted to establish a dual-color-ratio-feature spatial model. The clustering method based on the Gaussian mixture model is used to analyze deeply the features of migration and separability of the dual-color-ratio-feature space. The migration vectors of feature sub-space of different maneuvering modes and the change of the cosine of adjacent vector angle are obtained. In addition, the smallest attitude angle change of feature sub-space and the separable distance threshold are obtained as Δα=6.25° and Dth=2.6, respectively. It provides the basis and feasibility for the recognition of point target maneuvering modes. According to the characteristics of dual-color-ratio-feature sub-space, a method for the recognition of point target maneuvering modes based on sequential-feature sub-space is proposed, which is verified by simulation as simple and feasible as well as possesses high sensitivity and separability in the recognition of point target maneuvering modes. These results have great significance for the acquisition of point target maneuvering information in the beyond-visual-range air combat.

The study gets a closed form of the phase-response of a waveguide Bragg grating (WBG) by solving its coupled-mode equation with the Fourier transform (FT) of its index perturbation and the law of flux conservation, and then establishes the semi-analytic general solution of its delay spectrum by differentiating the phase response. Based on this delay general solution, the delay spectra of uniform and linearly-chirped WBGs are simulated, which are compared with those delay spectra obtained by other methods and the measured spectra in order to verify the analysis precision and efficiency of delay general solution. The comparison results show that the delay spectra calculated with this general solution agree well with those measured or calculated by other methods in the whole reflection band. Moreover, this general solution can be employed for the fast and exact analysis of arbitrarily complicated delay spectra of WBGs. The WBGs with analytic FT and discrete FT possess the linear complexities of O(N), and O(Nlb N)(N is the number of calculation points), respectively. This method may provide a universal basic theory and an analytic method for the analysis, design, and application of the delay properties and phases of WBGs.

In the coherent optical communication system, the traditional blind phase search (BPS) algorithm has shown its excellent tolerance to laser linewidth, however its high-computational-complexity limits its practical applications. Thus a low-computational-complexity carrier phase estimation (CPE) algorithm is proposed, in which the phase-test range is defined based on the slow variance characteristic of noise phase produced by laser linewidth and the optimum phase angle is searched. Consequently, the number of phase searches is finally reduced and the computational complexity of this algorithm is also significantly reduced. Meanwhile, the proposed algorithm can avoid the problem of few phase mismatches existing in the traditional BPS algorithm and thus it is superior to the traditional BPS algorithm in performances. Finally, the correctness and validity of this algorithm are verified in the polarization multiplexing coherent optical communication system.

A high-resolution transverse load fiber sensor based on microwave photonic filter (MPF) is proposed. It is different from the conventional fiber sensors based on spectral analysis because the signal demodulation is finished in the microwave domain. The working principle is as follows. With the polarization-maintaining fiber Bragg grating (PMFBG), a dual-wavelength ring fiber laser with a stable polarization is constructed. The microwave signal is modulated on the laser output and a MPF with two taps are formed with the time delay introduced by the long distance fiber. The theoretical and experimental results show that there exists a linear relationship between the frequency response of this filter and the transverse force on the PMFBG. The transverse force to be measured can be retrieved by the detection of the resonance frequency shift of this filter. In addition, in the experiment, a high sensitivity of 9.87 MHz·N-1 is obtained. The consistency between the experimental and theoretical results confirms the feasibility of this method.

High-speed and large-capacity all-optical signal processing can be achieved in the future with the optical parallel logic operators. Such devices can help reduce data transmission delays. This study proposes a hybrid operator of three-channel optical phase for optical computing (A+B-C, A+C-B, and B+C-A) that is designed according to the phase-insensitive amplification principle of four-wave mixing (FWM) in fibers. The nonlinear coupled-mode equations for the cascade FWM are derived. Results reveal that between each output idler and input signal have a certain phase relation, which provides the theoretical basis for phase compensation. Moreover, parallel hybrid operators have amplitude noise index and phase noise transfer coefficient in terms of error vector amplitude (EVM) of 0.9 dB and 1.67, respectively. For a quadrature phase shift keying (QPSK) signal, when the signal-to-noise ratio is >24 dB and the EVM is <12%, the symbol error rate of error-free coding is <10 -3. Three- and single-channel operators exhibit the same phase noise transfer characteristics due to common FWM-phase matching conditions. However, the three-channel operator exhibits lower phase noise of 0.2 dB and larger power transfer efficiency that is more than twice that of the single-channel operator.

This study aims to propose a nanodisk structure embedding a rectangular metal block. The Fabry-Perot cavity formed by this structure is used to enhance the coupling effect of the surface plasmons. The structure has a narrow bandwidth, high quality factor, and high filtering performance. Herein, a multi-channel wavelength-division multiplexer is constructed by multiple cavities coupling. The influence of the horizontal and vertical widths of the rectangular metal block and the coupling distances between the embedded disk and rectangular metal block on the transmission characteristics of the device is described with the time-domain finite-difference method, for which a device without embedded rectangular metal block is used as the control group. A multi-channel wavelength division multiplexer is realized according to its transmission characteristics. The filter shows strong transmission characteristics when the disk resonance filter is embedded in the rectangular metal block; its full width at half maximum is significantly reduced and the quality factor is increased. By coupling a number of inline rectangular-metal-block/disk resonators, we construct the filter. Such plasmon multi-channel wavelength-division multiplexers can provide two- and three-channel demultiplexing functions. The resonant wavelength of each channel can be adjusted by selection of the parameters of the embedded metal block in the resonator, the transmission efficiency can reach up to 70%, and the minimum insertion loss is 1.549 dB. The average operating range is 189 nm, and there is no adjacent-channel crosstalk. We demonstrate that the proposed structure has good de-multiplexing frequency characteristics.

A novel temperature and strain dual-parameter fiber sensor is proposed, which is constructed via the cascading between a hollow core fiber and a fiber Bragg grating. The hollow core fiber confines the light to transmit inside the air core based on the anti-resonance mechanism and the light satisfying the resonance condition leaks out of the air core, which is indicated as a periodic loss peak in the transmission spectrum. Because the physical mechanisms for hollow core fiber and fiber Bragg grating as well as their spectral responses to external temperature and strain are different, the simultaneous measurement of temperature and strain can be realized based on the coupling matrix. The experimental results show that the temperature sensitivities of hollow core fiber and fiber Bragg grating are 24.55 pm /℃ and 12.76 pm /℃ near 1550 nm wavelength, respectively. In contrast, the strain sensitivities are -0.70 pm/με and 1.02 pm/με, respectively. The proposed dual-parameter sensor is simple in fabrication and has a high measurement accuracy.

A method for the automatic phase-distortion compensation in digital holography is proposed, in which the image segmentation technique is used to automatically segment the detected objects, thus a phase mask is generated, and the phase-distortion in areas without containing the detected object region is further obtained. The least-squares fitting of phase-distortion based on the phase-distortion correction model is done and the automatic phase-distortion compensation is eventually achieved. In experiments, a built-in digital holographic inspection platform is established and the wafer surface measurement is conducted. The results show that the proposed method can be used to automatically correct phase-distortion.

In order to avoid the details blurring and noise amplification, the image is decomposed as structural layer and texture layer, and the dehazing operation is only performed on the structural layer. The transmission fusion method based on the idea of frequency domain filtering is proposed to remove the block effects in the transmission image and the halo artifacts in the restored image. To solve the problems existed in the transmission optimization process such as the complex computation, being incapable of keeping the balance between the transmission smoothing and details preservation, we propose the multi-guided filtering method. At the same time, an adaptive atmospheric light estimation method is proposed which can be applicable to the scenes with large white objects. Experimental results show that the proposed algorithm can remove the haze effectively and the restored image has clear details and natural color. The noise and halo artifacts are suppressed remarkably, and the contrast and saturation of the scene are improved significantly.

Aiming at the specific application of building detection, a sparse representation-based full-reference quality assessment method for distorted satellite stereoscopic images is proposed. First, a new distorted satellite stereo image database is constructed, in which the corner detection and the digital surface model elevation information are used for building detection. And a detection accuracy index is proposed to represent the degree of distortion based on the change of the detected corners. Then, an objective evaluation model based on sparse representation is proposed, which extracts scale-invariant feature transforms and binary robust invariant scalable key points of the original and the distorted images for dictionary learning. Four quality scores are obtained using sparse representation to measure the similarity between the original and the distorted images. Finally, the final objective assessment value is obtained by fusing the four quality scores using support vector regression. The test is carried out on the constructed database. The test results on the constructed database show that the Pearson linear correlation coefficient is higher than 0.90, and the Spearman rank correlation coefficient is higher than 0.87. Compared with the existing assessment methods, the proposed objective evaluation method can better reflect the quality of satellite stereo images.

Calculation of nominal fixed grid, aimed to project a given area of the earth to the nominal image, is a key technique for navigation and registration of geostationary satellites. In view of the existing nominal fixed grid definitions from coordination group for meteorological satellites (CGMS) specification and geostationary operational environmental satellite-R series (GOES-R) user's guide, whose representatives are FY-4A, Himawari-8, MTG, Electro-L satellites and GOES-R satellite, respectively, the formulas of CGMS and GOES-R nominal fixed grid calculation are deduced in detail, and the difference and relationship between CGMS and GOES-R nominal grids are summarized. Both nominal fixed grids are defined in satellite body Cartesian coordinate system. Furthermore, the detailed transforming process from nominal image of CGMS specification to the corresponding nominal image of GOES-R definition is presented using FY-4A observations. The results demonstrate that the nominal images defined by the two aforementioned methods can be converted to each other. Besides, by comparing the scanning and stepping angles of FY-4A advanced geosynchronous radiation imager to the responding angles of GOES-R fixed grid, the differences are 10-16 μrad and 10 -17 μrad magnitude, respectively, which reveals that GOES-R fixed grid is defined from the point view of opto-mechanical design of advanced base imager. Considering that the foreign satellites, except GOES-R, all use CGMS fixed grid to calculate the nominal image, for end-user's convenience, FY-4A nominal fixed grid will adopt the internationally accepted standard projection defined in CGMS LRIT/HRIT global specification.

The computed laminography (CL) system has a unique advantage in aspects of large and plate-like objects imaging. We propose the parallel translation computed laminography (PTCL) system. Then, aiming at the image reconstruction of the system, the Feldkamp, Davis and Kress (FDK) algorithm is applied in the system. Due to the limited size of the detector, the system can only collect the projections of the region of interest of the object and the the total variation minimization based simultaneous algebraic reconstruction technique (SART+TV) algorithm is introduced into the object imaging. The simulation and experimental results demonstrate that both FDK and proposed method can achieve image reconstruction for PTCL. Compared with the FDK algorithm, the proposed method can reconstruct high-quality images from truncated and region of interest projections. Furtherly, it also demonstrates the feasibility of the system.

To meet the advanced technical requirements on X-ray streak camera in the inertial confinement fusion (ICF) experiments, a large format X-ray streak tube is designed and fabricated, in which a spherical phosphor screen is used to alleviate the influence of field curvature and also the increase of anode total voltage and deflection sensitivity is used for the improvement of time resolution. The test results show that the effective length of photocathode is 40 mm, the total static spatial resolution amount is 1000 lp, the dynamic spatial resolution is 20 lp/mm, the temporal resolution is 3.3 ps, and the dynamic range is 2281∶1, which can meet the capturing requirement of a large amount of information in the ICF experiments.

A mathematical model related to stage errors and grid plate errors is constructed according to the different positions of grid plate in the auxiliary measuring device. Based on the least square principle, the error equation is converted into a regular equation. The influence of position scheme on the rank of a relation matrix is investigated to summarize the law between the position and the self-calibration model. With the the conditions that the equation need for the least square solution, in the self-calibration process, the grid plate must be rotated by 90° and simultaneously translated from its initial position, and this position transformation is also simulated. The research results show that it is only the position scheme containing three basic positions that makes the simulation result closer to the true value. These three basic positions are the necessary and sufficient conditions for the achievement of self-calibration by the least-squares method.

A three-dimensional (3D) tracking registration method is proposed based on the semantic object matching. The improved single-shot multi-box detector (SSD) deep convolution neural network is used to segment images semantically and thus the pixel level semantic segmentation results for different objects in the scene are obtained. To solve the object function of the camera pose, the camera pose is estimated by the combination of the gray and the geometric constraints of images. The proposed method not only reduces the influence of the lack or mismatch of feature points on the performance of 3D tracking registration algorithm, but also it can adapt to the scenes with different structures. The research results show that the error of this proposed method is less than 2.2 pixel, which basically satisfies the requirement of real-time.

A space-borne far ultraviolet imaging spectrometer working at the sun-synchronous orbit is studied, which is used to satisfy the scientific requirements of the connection and coupling detection between the low atmospheric driver and ionosphere. The instrument is designed to detect the spectral intensities of various particles in the ionosphere, which produce the characteristic radiation different at day and night in the far ultraviolet band, by the two-side lateral limb observation method. Then the influence of the low atmospheric driver is further acquired quantitatively. According to the detection mechanism, the observation scheme for instrument system is designed, the instrument performance parameters and system design, integration of prototype system, performance test, ground radiation calibration, and others are investigated. This research provides a new way in the future far ultraviolet imaging spectral observation of the ionosphere.

Tunable diode laser absorption spectroscopy (TDLAS), as an accurate detection method for the trace gases, has been widely used in the life and production. It can accurately inverse the gas concentration through the linear relation between the integral absorbance and the gas concentration. The changes in the environment and the system noise tend to cause the deformation of the absorbance curve. Therefore, it is necessary to do the nonlinear curve fitting on the absorbance curve to make it return to the Voigt model. A real-time online monitoring system for carbon monoxide based on TDLAS has been designed and built. With the platform, a fast calculation method with the triangular Voigt linetype fitting for the single spectral integral absorbance has been put forward and compared with the Gauss-Hermite method. The accuracy of the proposed method is only 0.11% lower, but the average computation time is shortened by 84.19%. The experimental results show that the triangular Voigt linetype fitting method can greatly improve the fitting speed only with the minimal accuracy loss.

In order to improve the detection accuracy of soil elements by laser induced breakdown spectroscopy (LIBS), we establish a quantitative analysis model for soil elements of relevance vector machine (RVM). And it is compared with support vector machine (SVM) model and least squares support vector machine (LSSVM) model. The four characteristic lines of Ni element are taken as the analysis lines, after full spectral normalization, RVM, SVM and LSSVM models are established with the training sample set. According to the test results of testing sample sets, we can know that the SVM is inferior to the others model in terms of model prediction accuracy. However, in terms of model stability, LSSVM model is poorer than the others models. Therefore, in the practical applications, the advantages of RVM in model stability and prediction accuracy indicate that it is more suitable for quantitative analysis of laser-induced breakdown spectroscopy.

Herein, we use an optical phase locked loop (OPLL) for locking the local laser's phase to achieve better laser phase-lock and obtain highly coherent unit beams. First, the cross-correlation function of the phase-locked laser's complex amplitude is determined, and the coherence between different phase-locked lasers is evaluated with a cross-correlation function. The influences of several key parameters, such as the linewidth of the reference light, time constant of the OPLL system, and the difference values of the local laser's linewidth on the function's value are analyzed. The evaluation parameter M is used to characterize the influence of the difference values of the local laser's linewidth on the coherence of the phase-locked laser keeping other parameters constant. Furthermore, the influence of the difference values of the local laser's linewidth on the phase-locking effect is analyzed in detail, and the related formula for calculating a reasonable value range of the related parameter is obtained. Our results indicate that the time constant of the OPLL system must be designed on the basis of the linewidth interval of a local laser to achieve effective phase-locking.

In order to solve the problem of difficulty and low accuracy of registration between cross-scale data, we propose a cross-scale data registration method based on fractal dimension characterization. The discrete wavelet transform algorithm is used to decompose the raw data to obtain its approximation at each scale. Then the fractal dimension is adapted to characterize the scale parameter of the decomposed data. According to the fractal dimension, we pick out the scale-approximated data from different raw data, using iterative closest point algorithm for fine registration as a usual case. The calculation result of fractal dimension at different scales shows that the fractal dimension can represent scale parameters effectively. And the registration results indicate that the proposed registration method can achieve the registration of cross-scale effectively.

The calibration of multi-camera pose relationship is a very important step in the large dimension measurement system. A calibration model is established and a planar target with a filling circle array is adopted. By means of linear translation, the target is located within fields of view of different cameras at the same time, the coordinate value of the same target feature point in each camera coordinate system can be extracted. Based on the coordinate values, the position-pose relationship between the target coordinate system and the camera coordinate system can be calculated. Moreover, according to the linear transition of target, the position-pose relationship between two cameras is thus solved. It is verified by experiment that, as for the cameras with an installation distance of about 2300 mm, the calibration error is less than 0.002 mm. The whole calibration procedure is simple and fast, which can be applied to the on-site calibration by a large dimension measurement system in industry.

Aiming at the problem that the accuracy and real-time of multi-target detection in complex and large scenes are difficult to balance in the existing target detection algorithms, we imitate the human visual mechanism inspired by the convolution kernel shape of the deep neural network. The target detection framework——the single shot multi-box detection (SSD) based on deep learning is improved, and a multi-target detection framework adaptive perceive SSD is proposed, which is specially used for the multi-target detection in complex and large traffic scenes. A feature convolution kernel library composed of multi-form Gabor and color Gabor is designed. The optimal feature extraction convolution kernel group is trained and screened to replace the low-level convolution kernel group of the original network, and effectively improves the detection accuracy. A single image detection framework is combined with a convolution long-short-term memory network, and the temporal association of network frame-level information is realized by extracting the characteristic mapping between propagation frames with a bottleneck-long-term and short-term memory layer. And the calculation cost is reduced, and the tracking and identification of targets affected by the strong interference in the video are realized. An adaptive threshold strategy is added to reduce the rate of missing and false alarms. The experimental results show that compared with other target detection frameworks based on deep learning, the average accuracy of various target recognition is increased by 9%~16%, the average accuracy is increased by 14%~21%, the multi-target detection rate is increased by 21%~36%, and the detection frame rate reaches 32 frame·s-1, which achieves a balance between the accuracy and real-time performance of the algorithm and achieves better detection and recognition results.

The focusing accuracy of the excitation laser on the target surface has a great influence on the stability of the spectral signal in the laser-induced breakdown spectroscopy (LIBS) system. But it is difficult to pretreat the natural samples in some special application scenarios, such as the LIBS detection of sediments in the vicinity of deep-sea hydrothermal vents, to improve the focusing accuracy. It is significant to design a focusing system with fast speed and high precision for in-situ LIBS detection. In this research, the microscopic autofocus system combined with the image contrast evaluation method is applied to the LIBS detection system in order to improve the focusing accuracy, and the two-color laser diode assisted focusing method is used to expand the focusing range and accelerate the focusing speed. The experiments show that combining the two focusing methods can realize 2400 μm focusing range and 20 μm focusing accuracy, and can speed up the focusing process. The system proposed is hopeful to improve the LIBS signal quality and spectral acquisition efficiency in field detections.

To address the computer vision-based registration problems, this study proposes a new three-dimensional point cloud registration algorithm that combines fast point feature histogram (FPFH) feature description with Delaunay triangulation. First, the FPFH is used to comprehensively describe feature information; then, the local correlation of feature information is established using the Delaunay triangulation. Thereafter, according to the corresponding relation of point pair features, the initial conversion of the sampling consistency is performed to implement initial registration. Finally, the iterative closest point method based on the initial values is used for accurate registration to obtain a precise conversion relation. The registration experiments are conducted on simple and complex target objects. Results reveal that traditional point cloud registration can be improved by combining FPFH feature description and Delaunay triangulation. This registration simplifies the feature extraction complexity, reduces the search range of matching feature points, improves the registration speed and accuracy, achieves an efficient registration of target objects, and considerably improves the efficiency of matching feature points in machine vision.

In the existing weight-adaptive cross-scale algorithms, the same weight for the cost function of each scale is adopted, the different influence of each scale layer on the whole matching cost is missing, and thus the number of mismatching points increases. As for this problem, a weight-adaptive cross-scale algorithm framework for stereo matching is proposed. The cost matching is performed on different scales in the framework of unified cost aggregation function and the information entropy of each pixel window is used as the influence factor of the matching cost at each scale on the whole matching cost. At the same time, a regularization factor is added to the cost function to ensure the cost consistency at different scales for the same pixel. The proposed algorithm framework can be applied to the multi-scale algorithm of cost matching and improve the accuracy and robustness of the existing algorithms. Based on the proposed algorithm framework, the different cost aggregate functions are tested on the Middlebury dataset. To ensure the fairness of tests, as for each algorithm, there is no a subsequent parallax refinement step. The experimental results show that the proposed algorithm effectively improves the accuracy and robustness of multi-scale stereo matching.

Circular control points detection is usually considered as the detection of ellipses. Due to the illumination, measuring angle and other reasons, the edges of control points may be incomplete. Furthermore, the noise edge of complex background or object will interfere with the extraction of control points. For those reasons, we propose a method of circular control point detection based on the circumscribed rectangle of an ellipse. Firstly, the center position of an ellipse is obtained by fitting director circle. Then the orientation and two semi-axis of an ellipse are determined by using the geometric properties of the external rectangle. The detected ellipses will be verified in order to reduce the detection errors. Finally, a clustering algorithm based method is proposed to extract the efficient control points. The simulation and real experimental results show that the proposed algorithm has high accuracy and excellent detection performance even for incomplete ellipses or cases in complex situation.

In air-to-ground remote sensing detection, the object has the characteristics of small field of view and single viewing angle, which is susceptible to background interference. At the same time, the height of the field of view varies greatly, which brings challenges to the traditional deep learning detection algorithm. To solve the problem, a scene-coupled multi-task object detection algorithm is proposed. First, a new scene-coupled object detection network structure is designed, which mirrors and fuses the scene classification feature map and the object detection feature map on the same scale to enrich the fine-grain of the feature description. Second, a differentiated activation module is designed to realize the importance screening of feature channels. Then, the optimization function of multi-task coupling is derived, which can simultaneously optimize the scene classification loss and object detection loss. Finally, an air-to-ground detection multi-task dataset is established to verify the effectiveness of proposed method. The experimental results show that the proposed algorithm effectively improves the accuracy and robustness of air-to-ground small object detection, and can adapt to different heights to identify multi-task requirements, which provides a new idea and method for space-based unmanned platform intelligent detection.

Due to the impact of particles, the mechanical agitation and other factors, the rising floatation bubbles causes severe deflection and deformation. The horizontal set partitioning method is proposed for measuring the volume and surface area of the air bubbles. First, we establish an observation system for moving bubbles during the flotation process, and collect the bubble image. We use an edge detection method based on area segmentation to extract bubble edge. For overlapping bubbles, we use curvature scale space corner detection algorithm and direction chain code to mark the pits, thereby divide the overlapping contour. The edges of independent bubbles are fitted and reconstructed by least squares. Then we calculate the deflection angle of the bubble according to the edge, and adaptively select the separation interval, and finally the volume and surface area of the bubble are obtained by the accumulation of the divided portions. Experimental results show that the edge extracted by this method is accurate and not easily affected by the light environment. Under the conditions of different agitation rates, the average error and standard deviation of the measured bubble volume are 4.52% and 0.057 mm3, which are more accurate than other methods.

The macroscopic calibration method cannot be applied in the microscopic three-dimensional digital image correlation system due to the small depth of field and the complex optical paths of a stereo microscope. As for this problem, a fixed-point rotation calibration method is proposed, which is suitable for microscopic stereo vision. The mathematical model is established based on the relationship between magnification and depth of field, and thus the maximum angle between calibration plate and XY plane is obtained. In addition, the fixed-point rotation platform is designed to calibrate the calibration plate. The calibration parameters are optimized by a series of experiments and the inclination angle used for minimizing the overall error of the calibration parameters is obtained. The calibration results show that the main point coordinate error is not larger than 1.8 pixel, the maximum deviation of the relative translation vector of Z component is less than 0.15 mm, and the calibration result tends to be stable when the attitude number is 10 or above. The accuracy of the calibration results is tested by use of a precision displacement platform and the results show that the average relative error of displacement measurement by the proposed method is 2.2% and the mean square error is 0.36 μm.

The polarization control scheme is generally used for generating attosecond pulse, in which the high harmonic emission only occurs within a half optical period of polarization gating and thus the obtained harmonic spectrum is continuous in the whole plateau and at the cutoff position. Based on the Ammosov-Delone-Krainov tunneling ionization theory and the strong field approximation method, the effect of the intensity ratio of two-beam counter-rotating circularly polarized pulses in the polarization control scheme on atomic ionization probability and harmonic emission power spectra is investigated by numerical simulation. It is found that if the intensity ratio of two-beam pulses is properly controlled, the effective atomic ionization can take place in the first quarter of an optical period of polarization gating, which is beneficial to obtain an optimal high harmonic spectrum. Moreover, one can optimize the cutoff position and conversion efficiency of harmonic spectra according to the needs, provided the intensity ratio is always less than 1.

An athermalization design method of a visible light optical system is proposed based on the grouping by weight. For this optical system satisfying the requirements of image quality at normal temperature, the comprehensive weight of chromatic power and thermal power produced by each element is calculated. The structures are grouped by the comparison among the comprehensive weights, and the initial system is equivalent to two single lens systems for the choice of the optical materials. The athermalization design of a double Gaussian aerial camera with a focal length of 400 mm and an F number of 4 is conducted, whose modulation transfer function (MTF) is larger than 0.4 at a spatial frequency of 55 lp·mm-1 under the temperature in the range of -40-+60 ℃. The research results show that this design method can ensure the telephoto optical system possess good image quality and stable optical performance under the environment of large temperature difference.

This study proposes a fast simulation method that employs machine learning-based parameter calibration for three-dimensional (3D) rigorous simulation of defective masks in extreme ultraviolet lithography. The parameters of the structure-decomposed fast simulation model for defective mask diffraction are calibrated using machine learning methods, such as random forest and K-nearest neighbors, to improve the simulation accuracy and adaptivity. Herein, rigorous simulation is used as a benchmark standard for the calibration of model parameters. Simulation results of 50 validation contact masks set randomly reveal that the average simulation accuracy of aerial images is increased by 45% after calibration; both calibrated and uncalibrated fast models display better simulation accuracy (improved by 4.3 and 8.7 times, respectively) compared with an advanced single-surface approximation model. By applying defect-compensation simulation to a mask of 44-nm period, the simulation speed of single diffraction of the corrected fast model is ~97 times faster than that of the rigorous simulation when the simulation results are consistent (error is 0.8 nm).

The viscosity of the polishing slurry is an important parameter during the fluid jet polishing (FJP) of optical elements. Polishing slurry viscosity in the FJP of hard and brittle optical components has not been extensively investigated. This study therefore investigates the influence of viscosity on the material removal function in the FJP process. The physical models of continuous-phase, discrete-phase, and erosion processes are also proposed for FJP. Pathlines of abrasive particles with different viscosities are calculated, and the variation rule of impact velocity vector of abrasive particles with viscosity is analyzed. Using the polishing slurries with different viscosities but the same mass fraction, the effect of viscosity on material removal function is obtained by combining the static mining-spot experiment with plastic-wear theory for a BK7 optical glass workpiece. The influence of viscosity on the material removal function and the influence on the surface roughness after uniform polishing are evaluated. The results indicate that as the slurry viscosity increases, the form of the material removal function and the material removal range remain the same, whereas the material removal depth decreases exponentially and the surface roughness of the optical elements is improved. Results of this study expand the knowledge of FJP material removal mechanisms and have theoretical significance regarding the control of polishing slurry viscosity.

A novel photonic crystal (PC) structure with periodic rectangular holes is proposed to realize the narrow-band polarization notch filtering of visible lights. This PC structure is equivalent to a periodic structure with a dielectric-grating-dielectric layer in the PC structure model. The PC structure is first analyzed by the Rugate filtering theory and then the equivalent medium theory (EMT) and the transmission matrix method (TMM) are combined to simulate the transmittances of lights with s and p polarizations. The effects of the parameters, such as longitudinal period number m, filling ratio of grating layers f and thickness d of this PC structure, on the central wavelength, band width and transmittance in the cut-off zone for polarization notch filtering are also discussed. As for the central wavelengths of 417, 497, 582, 685 nm, a p-polarization notch filtering structure with a band width of 10 nm is designed, which is tested by the simulation with the finite difference time domain (FDTD) method. The results show that the PC structure with periodic rectangular holes can be used to realize the narrow-band polarization notch filtering of visible lights.

The heat transfer performances of a sunflower radiator with triangular groove extended surfaces for high-power LEDs are numerically simulated and analyzed. The temperature distribution in the direction of fin length is experimentally tested. Under the condition that the natural convection and radiation model is considered, the effects of apex angle α, groove width s and groove depth d on the maximum temperature rise ΔTmax at the top of fins, average convective heat transfer coefficient h and convective thermal resistance R are investigated. The results show that the existence of triangular grooves with apex angles of 90°-120° and inclining to fin root increases the heat dissipation area and also improves the fluid-flow distribution, and thus the heat transfer performances of sunflower radiator are significantly enhanced. Compared with that of groove width s, the influence of groove depth d on the average convective heat transfer coefficient h is more remarkable. A small or large groove depth deteriorates the heat transfer performance due to the significantly decreased average convective heat transfer coefficient h.

With the discrete phase encoding, a free space quantum key distribution system is constructed based on the phase synthetic polarization technique. On the basis of the temporal filtering, spatial filtering and spectral filtering, the polarization filtering combined with the model of the polarization degree of the atmospheric background light is used to suppress the background noise. The research results show that, the background light suppression ratio is up to 5 when the polarization degree of the atmospheric background light is 0.6. Compared with the situation without polarization filtering, the bit error rate of the system caused by the background light is reduced by 80%.

Based on the Kolmogorov and non-Kolmogorov turbulence models, the scattering effect of atmospheric turbulence on orbital angular momentum (OAM) is analyzed and the probabilities of different OAM modes are obtained at the detecting end. The key generation rates and the maximum propagation distances of OAM-encoded measurement-device-independent quantum key distributions (MDI-QKD) under two atmospheric turbulence conditions are analyzed. The simulation results show that, with the increase of the radial intensity during the light beams are propagating in the atmospheric signal channels, the scattering effect of the turbulence on OAM is gradually enhanced, and the initial OAM states are gradually diverted to the adjacent modes with a tendency of random distribution. The probabilities of initial OAM states at the detecting end decrease gradually. The maximum propagation distance for the OAM-encoded MDI-QKD is about 10 km longer than that for the polarization-encoded MDI-QKD under the atmospheric turbulence.

Compared with the traditional laser altimetry technology, the photon-counting laser altimetry technology has the advantages of large data, light weight, and high ranging precision, which is the development trend of laser altimeter technolgoy. In this paper, we establish a mathematical model to study the characteristics of the photon-counting laser altimetry. The performance of the photon-counting laser altimeter is estimated by numerical calculation. The ground object model is established, and the simulation is carried out with Monte Carlo method. A filtering method for the altimetry data and an algorithm for optimizing the elevation information using the remote sensing images are proposed. The results show that the root-mean-square error of the photon-counting laser altimeter is 6.1 cm under the condition of the noonday background with the most intense sun for the typical ground model, and the error after optimization by the algorithm is 2.6 cm.

A stripe-removing algorithm using low-pass filtered residual images is proposed herein to remove stripe noise in hyperspectral remote sensing images. First, a Gaussian low-pass filter is used for image filtering to obtain a low-pass filtered residual image. Then, using previously determined knowledge that the rank of the stripe noise is 1 and the details are orthogonal to the stripe noise, we employ the orthogonal subspace projection technique to separate the stripe noise from the details in a low-pass filtered residual image. Finally, the separated details are then added to the filtered image. Through continuous iteration of the above mentioned three steps, the proposed algorithm can effectively remove stripe noise and overcome image blurring issues caused by traditional low-pass filtering methods. The experimental results illustrate that the proposed algorithm can significantly improve the removal of stripe noise and preserve image information comparing with the existing stripe-removing algorithms.

The third and fourth harmonic outputs at 354.7 nm and 266.0 nm for Nd∶YAG pulsed laser are both within the ultraviolet wavelength band, and can be selected as the pump sources of a water vapor Raman lidar. Based on the practical construction of a water vapor Raman lidar system and from the aspects of backscattering coefficients, extinction coefficients, atmospheric transmissivity, ozone absorption and background noises, the detection performances of this lidar system are analyzed. The effects of the 354.7 nm and 266.0 nm laser sources on the daytime water vapor detection by this Raman lidar are investigated. The results show that the 266.0 nm laser and its corresponding vibrational Raman wavelengths of oxygen, nitrogen, and water vapor are all within the blind ultraviolet region, not affected by the solar background noise but affected by the ozone absorption. At 354.7 nm, there is no ozone absorption, but it is still influenced by the solar background noises. When the parameters for this wavelength selector system are kept the same, the effective detection range of daytime water vapor obtained by the designed Raman lidar is 2.7 km when the 266.0 nm is selected as the transmitter wavelength. In contrast, if 354.7 nm is selected, the effective detection range is only 0.6 km. The solar blind ultraviolet wavelength selection can effectively increase the detection range of daytime water vapor by Raman lidar and ensure the realization of all-day water vapor detection.

There exist two factors influencing the accuracy of conventional hyperspectral target detection. One is the inherent image noises induced by spectral distortion, and the other is the equal contributions of all adjacent pixels in the heterogeneous region. However, in fact the heterogeneity implies that the pixels are composed of different materials and possess different spectral characteristics. To address these problems, we propose a hyperspectral target detection method by the combination of spatially adaptive model and sparse representation. The noise sparse representation is utilized to reconstruct an accurate signal, in which the useful information in noises is extracted as possible to make the reconstructed signal be full of more features and be close to the original signal. In addition, a spatially adaptive weighted model is proposed to detect the similarity between central pixel and neighboring pixels, and to make full use of the relationship among neighboring pixels. The final experimental results show that the proposed method possesses a strong robustness compared with the conventional hyperspectral target detection methods.

The existence of clouds in the atmosphere degrades the accuracy of aerosol retrieval. The empirical threshold method is popular in could detection, however its strong subjectivity and difficulty in coping with the dynamic spatial-temporal changes of the environment or the difference among satellite-borne detectors result in a large classification error at the boundary of ‘cloud’ and ‘clear’. In addition, its automatic detection is also poor. To achieve an effective detection of cloud over the land surface in the atmosphere, we propose a threshold optimized method which combines the statistical classification with data fusion of polarized multichannel remote sensing images. As for this method, a dual-brightness threshold to distinguish ‘cloud’ from ‘clear’ for most pixels is first derived based on the semi-supervised Kmeans clustering and its statistical features. Then, the joint confidence factor of multichannel data is calculated by the D-S evidence theory for each pixel in the fuzzy area of threshold neighborhood, and thus the fine threshold is acquired. The two objects of ‘cloud’ and ‘clear’ are finally and accurately classified in the sequential decision process. To validate the effectiveness of the proposed method, we perform a cloud detection experiment based on the remote sensing load data of POLRED3, and compare the measured results with the results of POLRED3. The results show that the classification by the proposed method is well consistent with that by the POLDER method with a high conformity of 95%. The error pixels are mostly located at the boundary between cloud and clear, indicating that the proposed method exhibits a favorable sensitivity to the classification at the cloud edge.

To simplify the temperature extraction steps for Brillouin scattering and also improve the extraction precision, we propose a new method for directly obtaining the temperature characteristics of Brillouin gain spectra based on the radial basis function neural network. The Brillouin scattering spectra at various temperatures are used as the training set to establish the temperature model. The temperature can be obtained through directly inputting the Brillouin spectra into the model. The effects of three methods of smooth fitting, back propagation neural network and radial basis function neural network on the temperature measurements are compared. In the experiment, 77 groups of data at sweeping frequency intervals of 0.175, 1, 5, 10, and 20 MHz are selected and also those at different linewidths are expanded. The results show that, the root-mean-square error (RMSE) based on the radial basis function neural network is relatively small. Moreover, the RMSE increases slowly with the increase of step frequency. When the step frequency is 20 MHz, the error of single line width is up to 0.8002 ℃ and that of multiple line width is 1.0814 ℃, 33.04% and 42.88% of that by the smooth fitting method, and 40.25% and 55.89% of that by the back propagation neural network, respectively. The convergence is improved to a certain extent as a result of calculation step reduction in the method based on the radial basis function neural network.

An information feedback deviation-weighted method based on light intensity autocorrelation function (ACF) reconstruction is proposed to make full use of the effective particle size distribution (PSD) information in the decay section of ACF. The deviation-weighted inversion is carried out successively and the next deviation is reduced until the defined minimum information deviation is reached, that is, the distribution-reconstructed ACF obtained by inversion tends to be consistent with that obtained from the photon correlator. The inversion of the simulated data of the broad distribution and closely spaced bimodal distribution granular system at different noise levels is conducted. The results show that, compared with the routine weighting inversion methods, the proposed method can be used to obtain more accurate inversion results for the broad distribution and the closely spaced bimodal distribution and a better anti-noise performance is demonstrated, which are verified by the inversion results of the actual measurement data of standard polystyrene latex particles.

The concentration changes of exhaled carbon dioxide and water vapor are closely related to the physical condition. Therefore, it is of great significance to detect their concentrations. An exhaled gas detection device based on a 2.73 μm distributed feedback laser is proposed. Spectral lines at 3659.402 cm -1 and 3659.934 cm-1 are selected to measure carbon dioxide and water vapor respectively by using the wavelength modulation spectroscopy. The results show that using the second harmonic signal to calibrate the gas concentration, the linearity of 0.99945 and 0.99679 is obtained when the volume fractions of carbon dioxide and water vapor are less than 35% and 2.3%, respectively. The concentrations of carbon dioxide and water vapor during the respiratory cycle are measured in real time. With the measurement time of 0.92 s, the sensor achieves a detection sensitivity of 4.33×10-3 and 1.37×10-4 for carbon dioxide and water vapor, respectively. At the acquisition time of 56.8 s, the detection accuracy with 0.12% of carbon dioxide is achieved, and a detection limit of 1.49×10-4 at the optimal integration time of 17 min is achieved for carbon dioxide measurement.

Micro energy dispersive X-ray diffraction (EDXRD) analysis has importance application prospects in measuring the phase structure of small samples or sample micro-area. A novel method of micro EDXRD analysis using a self-development micro X-ray fluorescence spectrometer is proposed. The two-dimensional scan of a 4 mm×4 mm mico-area on the surface of a RMB-5-Jiao coin is made by a portable micro-beam X-ray fluorescence spectrometer based on polycapillary X-ray optics focusing (focal spot diameter is 190.7 μm). After the obtained data are processed, the mappings of Cu3Sn(0 8 3), CuO(2 0 2) and other crystal phases are presented. Simultaneously, the two-dimensional scan of an ore particle with diameter of about 1 mm is also made by a desktop micro-beam X-ray fluorescence spectrometer based on polycapillary X-ray optics focusing (focal spot diameter is 31 μm). The mappings of SiO2(3 2 9), Fe2O3(1 1 6) and other crystal phases are also presented. The results show that the micro-beam X-ray fluorescence spectrometer based on polycapillary X-ray optics focusing has a potential application in the micro energy dispersion X-ray diffraction analysis of small samples or micro-areas of samples.

Based on the infrared absorption spectrum of methane, a dual-bandpass filter is designed by the spectral splitting. With the adjustment of ion source beams, the refractive index of the ZnS film layer is improved and the background noise is eliminated. The fabrication difficulty of aperiodic narrow-bandpass filter films is reduced by the introduction of controllable thin film layers. The experimental test and analysis show that the prepared dual-bandpass filter possesses the transmissivity of 89.25% at 3.31 μm and 81.97% at 7.67 μm, respectively, and the goal for improving the sensitivity of detecting systems is achieved.