To investigate the laws of atmospheric background radiation and instrument radiation as well as control the instrument thermal radiation and instrument accuracy, a measurement system of atmospheric background radiation is designed. The effects of various error sources, such as each component of this measurement system and radiation calibration, on the total measurement error are analyzed. Meanwhile, the factors influencing the accuracy of calibration are also analyzed. Finally, the direction of improving the measurement work is determined. These results show that the measurement error of this system in the calibration region is mainly composed of the radiometric calibration error and random error, whose values are 2.4719% and 0.0790%, respectively. In addition, the composite error is 2.4732%. For most of excellent astronomy sites, the atmospheric radiation intensity is far lower than the calibration value, and thus the extrapolation measurement is unavoidable. The estimation results of the extrapolation measurement errors indicate that the extrapolation measurement can lead to relatively high measurement errors. Therefore, in order to improve the precision of atmospheric background radiation measurement, the standard radiation sources with low radiation intensity are indispensable. The development of the atmospheric background radiation measurement system and the measurement work in the field are helpful to experience accumulation for the further development of large aperture infrared astronomical telescope systems and their practical applications to infrared astronomical observations.

A global and local deep feature based (GLDFB) bag-of-visual-words (BoVW) model is proposed. The high-level features extracted from the deep convolutional neural network are reorganized and encoded by the BoVW model and the fusion features are classified by the support vector machine. The features from the convolutional layer containing the local details and the fully-connected layer containing the global information of scenes are fully used and thus the efficient expressions of the remote sensing image scenes are formed. The experimental results on two remote sensing image scene datasets with different scales show that, compared with the existing methods, the proposed method possesses unique advantages in the representation ability and the classification accuracy of high-level features.

The performance of mixed radio frequency (RF)/free space optical (FSO) airborne communication system based on decoding and 2×2 relaying is analyzed. The model of 2×2 relay-assisted mixed RF/FSO airborne communication system is established, and the probability distribution function and cumulative distribution function about signal-to-noise ratio of this system are derived by Meijer’s G function. Moreover, the closed expressions of average bit-error-rate (BER) and outage probability of the system are obtained. The effects of atmosphere turbulence intensity, aperture size and modulation mode on the average BER and outage probability are analyzed by the simulation. The results show that the aperture averaging effect can effectively improve the performance of mixed RF/FSO airborne communication system, and the performance of the 2×2 relay-assisted communication system is significantly better than that of the 1×1 relay-assisted communication system.

A pure rotational Raman lidar for measuring atmospheric temperature on daytime is proposed. The quadruple frequency of 266.0 nm of Nd…YAG pulsed laser is selected as the transmitter wavelength, which avoids the effect of solar background light on the detection of atmospheric temperature. A new type of triple-diffraction double grating polychromator is designed as the spectroscopic system of lidar. Simulation results show that the designed double grating polychromator can achieve the extraction of Stokes and Anti-Stokes pure rotational Raman signals from high and low quantum number channels. The suppression rate of the Mie-Rayleigh scattering signals is up to 60-70 dB. Meanwhile, considering the effects of the ozone absorption and fluorescence on the detection performance of the system, the designed solar-blind ultraviolet pure rotational Raman lidar system with the analog detection mode can detect atmospheric temperature in the range of 2.2 km in 12 minutes' integration time.

Based on the different ability for detecting haze of each spectral band of remote sensing multispectral image, the blue spectral band which is more sensitive to haze is used to label the haze and non-haze images. The red spectral band which is less sensitive to the haze and the mean-shift segmentation are used to label bright object. Bright object can correct and compensate the haze images before dehazing processing. Taking domestic high-resolution remote sensing image as test object, the correction effect is evaluated based on the quality parameters of high-resolution remote sensing images. The research result shows that all image quality parameters of each spectral band are improved after correction, and image features in non-haze area keep consistency before and after correction. Detailed information of image in haze area is recovered and image quality is significantly improved after correction.

Taking LYTRO camera as an example, we propose a super-resolution algorithm for light field image with precise color vector. According to the hexagonal distribution of the microlens array inside the camera, combining with the point spread function and the arrangement of the filters on the camera detector array, we calculate the RGB color component values of each single pixel point, and accurately recover the color information of each pixel. The timeliness of color restoration is optimized by using pyramid algorithm. In the second part, a super-resolution algorithm based on color vector constraint of camera sub-aperture image sequence is proposed to improve image quality. The proposed color restoration method can be applied to many kinds of light field cameras. At the same time, the color vector is used as the constraint condition, and the color super-resolution image restoration effect is ideal.

Based on the characteristics of the noise source of airborne infrared search track (IRST) system, the detection probability model is established, and the effect of the threshold noise ratio (TNR) on the detection probability and the false alarm probability is analyzed in the detection process. The influences of the detection distance, target speed and combat environment on the detection probability are also analyzed based on the combat background. The dynamically setting of TNR in the process of detection is proposed and the specific solving method is presented. The value ranges of TNR are simulated under different mission requirements. With the solved TNR, the TNR-variable detection probability envelope is obtained. The research results show that compared with that by the method with a constant TNR, the detection probability envelope obtained by the TNR-variable method can adapt to the real-time change of air conditions, has a greater universality and a larger detection range.

On the basis of the equivalent-medium theory and the body phase delay of multi-layer diffractive optical elements, considering the influences of phase modulation caused by antireflection films, the surface microstructural parameters of multi-layer diffractive optical elements are optimized to realize the diffraction efficiency of 100% at the design wavelengths and the high polychromatic integral diffraction efficiency within a broad waveband. With this optimal design method, the diffraction efficiency and polychromatic integral diffraction efficiency of multi-layer diffractive optical elements with antireflection films working in visible wavebands are analyzed. The results show that with this optimal design method, the diffraction efficiency of 100% at the designed wavelengths and the high polychromatic integral diffraction efficiency within a broad waveband can be achieved under the premise of ensuring the physical effects of antireflection films. This method makes up for the traditional design defects of multi-layer diffractive optical elements and improves the design theory of multi-layer diffractive optical elements, which provides a reference for the design of hybrid imaging systems.

An improved Gerchberg-Saxton (G-S) algorithm is proposed to realize the accurate design of the diffractive optical elements based on annular beams. This algorithm ensures a small sampling interval on the output plane and suppresses speckle. Compared with the design results of the conventional improved G-S algorithm without speckle suppression, a uniform spot with better performance is obtained by the proposed algorithm. The simulated results are in agreement with the experimental results.

Based on the different deformations of optical fibers under different bending curvatures, a novel optical fiber curvature sensor is proposed and demonstrated based on seven-core fiber (SCF) and few-mode fiber (FMF). In the experiment, three kinds of sensors with different structures are prepared, and their interference modes are analyzed with the fast Fourier transform (FFT) method. The sensing sensitivities of wavelength are 6.33, 30.83, 15.96 nm/m, respectively, while the sensing sensitivities of intensity are 8.57, 25.65, 1.96 dB/m, respectively.

The CO2 laser point-by-point exposure technique is used to fabricate a long period fiber grating (LPFG) operating at 2 μm, and the transmission characteristics are also been experimentally explored. The influences of grating writing parameters including grating period, modulation depth of refractive index and number of grating period on characteristic loss peak at 2 μm waveband are studied. The simulated and experimental results show that the resonance center wavelength and resonance peak depth of LPFG at 2 μm are tuned by grating period and grating length, and laser scanning times and refractive index modulation will increase the coupling strength of the mode within the fiber. In addition, the temperature-sensitive property of a 2 μm LPFG is also explored and the sensitivity is measured to be 74 pm/℃ via the designed experiment. It is believed that this work possesses potential application value in the development of 2 μm fiber lasers and their key devices.

A novel automatic bias control algorithm based on in-phase quadrature-phase silicon-based optical modulator is theoretically and experimentally proposed. Firstly, we compare the modulation characteristics between silicon-based optical modulator and lithium niobate modulator. Based on the model of in-phase quadrature-phase silicon-based optical modulator, we derive the functional relationship between the output signal and the electric field signal of the modulator. A control scheme is proposed to add the sinusoidal dither signal into the bias operating point of the modulator and analyze feedback signal spectra of the low-speed monitoring photoelectric diode which is built-in modulator. The feasibility of the scheme is verified through numerical simulation. Then we build the experiment platform of automatic bias control. We obtain the test results of monitoring photoelectric diode signal spectra with different bias voltages applied to channel I (in-phase), channel Q (quadrature-phase), and channel Phase (outer-phase) by using the experiment platform. The model of the photoelectric transfer function described in this paper is supported by simulation and measurement results. Finally, we measure the control accuracy of the algorithm in 128 Gb/s dual-polarization quadrature phase shift keying modulation format and calculate the penalty factor of the optical signal-to-noise ratio by the vector magnitude error parameter.

A bandwidth tunable Mach-Zehnder interferometer (MZI) based on tapered-drawing fiber Bragg grating is presented. The two sides of this microfiber-assisted MZI are the tapered micro chirped long period fiber gratings (CLPG) , which are symmetrical about the central waist. The period and refraction index of the grating after fused tapering are modelled and simulated. The refractive index test results show that the relationship between the refractive index of NaCl solution and the reciprocal of passband width is linear. The transmission bandwidth can be tuned by changing the solution concentration. The adjustable accuracy is 0.64318 nm-1·RIU-1 obtained after the detection of the transmission bandwidth at 1550 nm.

In order to obtain clear wheat grain reconstruction slices, the conjugated ray is used to compensate for the missing data of Radon space. Based on the FDK (Feldkamp-Davis-Kress) reconstruction algorithm, the z-axis weight function is introduced to optimize the reconstruction results of the grain slice images. For the three-dimensional head model, the root mean squared error of the reconstruction result by the z-FDK algorithm is 3.6927 smaller than that of the FDK algorithm and the z-axis strength drop in the FDK algorithm is greatly suppressed. As for the projection images of wheat grains at egg stage, young larva stage and old larva stage, the experimental results show that the average gradient and the contrast noise ratio of the z-FDK algorithm are both larger than those of the FDK algorithm. The gray values of the reconstructed two regions of the hair end and the embryo end of wheat grains increase and the artifacts are improved.

Based on the mechanism of eyeball fretting, a model combining with the static and dynamic mechanisms is proposed, which is close to the non-classical receptive field model. In this model, two semi-elliptical rings on both sides of the receptive field axis are regarded as the non-classical receptive field suppression areas and the sub-regions are established as well. The fretting is simulated by suppressing the direction angle, and the final energy value is obtained by combining the suppression weight and the luminance features of the image. The experimental results show that using the proposed model to extract the target contours can restrain the background texture well and retain more contours, which is better than that of the traditional model.

When the digital micromirror device (DMD) is applied in the convergent path of the imaging optical system, the deflection movement of the micromirrors on the DMD surface around their respective rotation axes and the non-continuous distribution with respect to the plane mirror directly result in an optical path difference (OPD) between the off-axis chief ray and the on-axis chief ray. Based on the geometrical optics and the aberration theory, the relationship between the final OPD and the image height on the DMD surface, the incident angle, the DMD pixel size and the DMD tilt angle is obtained in the DMD-based convergent imaging path. The factors influencing the image quality of optical system and the compensation method are analyzed. The correctness of the theory is verified by simulation and experiments. The research results have a great significance to the design and adjustment of optical systems with DMD devices.

A model applied to the simulation of infrared targets is proposed. By the trained conditional deep convolutional generative adversarial networks, only the random noise and category label are necessary for the automatic generation of the simulation images of infrared targets belonging to the expected category. The parameters are trained on the handwritten digital dataset (MNIST) and the infrared dataset, respectively, and subsequently the automatic generation experiment is carried out, which can produce the high trueness sample images. The features extracted by the discrimination network are used in the classification experiments, and the images synthesized by the proposed method are used for data augmentation to improve the classifier performance. The research results show that the proposed method can effectively imitate the infrared radiation characteristics.

An automatic measurement method for corneal thickness of the optical coherence tomography (OCT) images is proposed based on boundary tracking, contour localization and curve fitting. The corneal thickness of the OCT images with high signal-to-noise ratio or with noise and artifacts is measured. The effectiveness and practicability of the proposed method are verified. The research results show that the proposed method not only can be used to accurately measure the corneal thickness of the corneal OCT images with high signal-to-noise ratio, but also has a good measurement effect for the corneal OCT images with noise and artifacts. Compared with the existing measurement methods of corneal thickness, the proposed method has the advantages of strong anti-interference ability and high measurement precision.

A searching method for an optimal code sequence of coded exposure is proposed based on the Memetic algorithm. The theoretical model for coded exposure imaging is analyzed and the criteria of fitness function for the optimal codeword selection is established. The Memetic algorithm framework is introduced to carry out the optimal code sequence search, and the genetic search algorithm is utilized to implement the global optimal solution search. On this basis, the simulated annealing algorithm is used to conduct the local optimal solution. The optimal codeword search results are obtained by the threshold constraint of the fitness function and the updated iteration of population and optimal solution. The research results show that, compared with other methods, the proposed algorithm can take into account both the global and the local optimal solutions, the obtained optimal code sequence has a better performance index, the execution efficiency is high, and the restored image has superior subjective and objective evaluation quality.

A full-field compensation method for nonlinear phase error is proposed. By measuring the reference plane for multiple times, the accuracy of phase unwrapping on the reference plane can be efficiently improved. An expected phase plane approaching to the ideal values can be extracted from these multiple measured phases on the reference plane. This expected phase plane can be used to detect the nonlinear phase error. When reconstructing the measured object, the nonlinear phase error can be directly searched in the look-up table with the unwrapped phase of the object, then the full-field unwrapping phase of the object be compensated. By using the proposed method to reconstruct several height-known flat planes, the maximum mean absolute error (MAE) is decreased from 0.48 mm to 0.06 mm and the maximum root-mean-square error (RMSE) is decreased from 0.55 mm to 0.07 mm.

A new method for precisely measuring mid-infrared waveplate phase retardation based on dual photoelastic frequency difference modulation is proposed. The modulation frequency of the waveplate measurement system is reduced by the frequency difference between two ZnSe dual-photoelastic modulators (PEMs), and low-frequency modulation signal carrying the tested retardation is generated. The phase retardation of the tested waveplate is obtained by dividing the 1-time frequency difference amplitude and 2-time frequency difference amplitude after modulation. The effect of light intensity fluctuations and PEM phase retardation fluctuations on the measurement precision of the system is effectively suppressed by the proposed method and the measurement precision is improved. The measurement principle is deduced theoretically. The ZnSe-PEMs and the experimental system are designed. The experimental results show that the relative error of phase retardation measured is better than 0.004% and the sensitivity can be up to 5×10-4 rad.

An application mode for high-precision path planning control application of automated guided vehicle (AGV) with a large space is proposed based on the combining of the workshop measurement positioning system (wMPS) and the fuzzy control algorithm. AGV path is planned in advance, and the real-time position and pose of AGV is accurately measured by wMPS, and the real-time forward and rotational speed parameters of AGV are output by the fuzzy control algorithm. Simulation and experimental results show that the positioning accuracy of AGV is better than 2.5 mm, which can ensure dynamic navigation and positioning accuracy of AGV.

A Jones-pupil-based method for measuring the polarization aberrations of the lithographic projection lens is proposed. A measurement equation based on the Jones matrix is derived, and a linear relationship between the intensity vector and the Kronecker product of the Jones matrix is built. The polarization aberration in the form of Jones pupil can be directly measured by this linear relationship. Using a Jones pupil of the typical lithographic projection lens as a measurement object, the numerical simulation are conducted to validate the proposed method. In these simulations, the actual parameter errors between the polarization elements and the charge coupled device (CCD) are considered. Measurement errors of this method are also compared with the results converted from the conventional Mueller pupil metrology. For the same rotation-angle configuration of polarization elements, the measurement errors of the diattenuation and the retardance by the proposed method are reduced significantly compared with that by the conventional method. The simulation results show that the proposed method can greatly improve the measurement accuracy of the polarization aberrations in the form of Jones pupil, without increasing the complexity of the existing metrology system.

In order to ensure the relative accuracy of measurement results of the imaging spectrometer, we must eliminate the polarization response of the optical devices in the spectrometer. A common method is adding depolarizer. The detection performance of the imaging spectrometer requires the depolarizer not only has the good depolarizing effect in response waveband of the spectrometer, but also affects the imaging quality of spectrometer as little as possible. In this paper, a compact double wedge depolarizer of limb imaging spectrometer for atmosphere remote sensing detection is studied. The wedge angle of the depolarizer is calculated and analyzed by means of integration, in addition, the parameters that affect the performance of the depolarizer are determined by MATLAB optimization simulation, and the related error analysis is carried out.

The structure of W band (2.73-4 mm) electro-optical phase modulator (EOM) is characterized by multi-parameters and complex design. Referring to the coplanar waveguide transmission line theory and EOM theory, and combining with the three-dimensional electromagnetic simulation high-frequency structure simulation software HFSS, we carry out the finite element analysis and calculation of the modulator. For key parameters in the geometry structure of the EOM, we carry out model simulation calculation in four dimensions, and establish the multi-dimensional mathematical model of the EOM. According to the simulation model, we obtain the variation rules of geometric parameters and main performance of the EOM, and establish the model structure of the EOM. Through structural optimization, we obtain the 100 GHz speed EOM model with matching speed, matching impedance, and small attenuation.

After the instability caused by lattice depth fluctuation is eliminated using power feedback, the effect of the phase fluctuation for a laser propagating through space or other media on the interference contrast of ultracold atoms in an optical lattice is investigated. A man-made and tunable phase noise is introduced by shaking the optical lattice and its effect on atomic interference contrast ratio is measured. It is found that the phase noises with different intensities have different effects on the experimental results. However, the phase noise with a small intensity has a trivial effect on the interference contrast ratio of ultracold atoms. At present, the measurement accuracy of interference contrast ratio cannot fully show the influence of laser phase on the coherent property of a cold atomic system. This means that this kind of optical lattice with a spatial configuration possesses a good robustness in the capture of ultracold atoms. Furthermore, in order to improve the phase stability, a laser phase locking technique is proposed for the optical lattice with a spatial configuration, which can effectively reduce the phase noise in the lattice system and provide a basis for the future high precision measurements by an optical lattice with a spatial configuration.

The picosecond-laser system at Shanghai Astronomical Observatory is analyzed and its performances, such as output powers and output waveforms, under different repetition rates are investigated. A stable laser output with repetition rate of 4 kHz, power of 3 W, and wavelength of 532 nm is realized. The 4 kHz repetition rate ranging system is established and the 4 kHz repetition rate satellite laser ranging is achieved. Compared with those for 1 kHz repetition rate, the amount of observation data and the precision of normal point data are increased by approximate 2.62 times and 1.62 times, respectively, which indicating the performance of satellite observation is improved. Moreover, the average spin rate of Ajisai satellite obtained by the Lomb algorithm is 0.4234 Hz with an accuracy of 0.0054 Hz, higher than those from the 1 kHz laser observation, and the measurement accuracy of satellite spin is also improved.

We propose a 10 kW fiber laser by using home-made large mode field gain fiber, passive components and dual-end-pumping technique to suppress nonlinear effects. A power amplifier with output power of 10.14 kW, wavelength of 1070.36 nm and 3 dB bandwidth of 5.32 nm is established. For the main amplification stage, the maximum optical-optical efficiency is 87.8% and the slope efficiency is up to 89.2%. To the best of our knowledge, the proposed laser is the most efficient 10 kW-level fiber laser reported in China.

A dense stereo matching algorithm is proposed based on image segmentation. This algorithm combines the gray-gradient algorithm and the zero-mean normalized cross-correlation (ZNCC) algorithm to generate matching cost. The SLIC (Simple Liner Iterative Cluster) algorithm is used for image segmentation. A method based disparity map and superpixels is proposed to update the matching cost. At the disparity post-processing stage, the LRC (Left Right Check), hole filling and cross adaptive window weighted median filtering methods are used to reduce the error matching rate of the disparity map. The performance evaluation experiments on four Middlebury stereo pairs demonstrate that the proposed algorithm achieves an average error matching rate of 4.99%.

Aiming at the status that the scanning laser surface of transmitter in the model of workshop measurement positioning system (wMPS) is regarded as an ideal plane, a kind method of the visual evaluation of laser surface is proposed based on a high-precision rotary table. On basis of the evaluation results, a linear folding judgment mechanism is introduced, and the mathematical model of wMPS scanning laser surface is optimized and reconstructed to reduce the systematic errors in the intersection measurement model. The new model is evaluated by the large-scale high-precision coordinate field. The results show that the new wMPS scanning laser surface model has a better effect in the fitting of an actual laser surface, and it is helpful to improve the measuring accuracy of wMPS.

In order to simultaneously detect the shadow and penumbra regions, this study proposes and proves the consistent properties of shadow region radiation. Point collection in contour (PCC) within the outline of the super-pixel region corresponding to moving object is obtained, and the PCC is divided into the object foreground area (foreground PCC) and the shadow area (shadow PCC). With the proposed region-based full shadow detection and object mask growth algorithm, the foreground object mask is inversely grown by combing the complete shadow region, the complete foreground region and the ViBe mask of the moving object. The experimental results in the public dataset show that the average accuracy of the shadow detection of the proposed method reaches 82.5%, and the performance is significantly better reaches the traditional method. The average growth rate of the object mask reaches 8.84%, and the accuracy rate reaches over 95%.

The perspective model is used to design the cooperative beacons for fixed wing unmanned aerial vehicles with visual in landing. The sensitivity of the beacon imaging to the guidance precision is analyzed, and the design method for the minimum distance between the beacon feature points is obtained. In the process of the UAV(Unmanned Aerial Vehicle) getting close to the beacon, the iterative calculation of the field of view boundary is carried out, and the design method of the layout range of feature points is given. The OpenGL visual simulation system is built to analyze the sensitivity of images and the capturing situation of beacon images under the 2th level guidance precision required by ICAO (International Civil Aviation Organization) and different positions and attitudes of UAV. The measurement results show that the proposed method can guarantee the design of cooperative beacons to meet the requirement of image sensitivity and the beacon can always be completely captured.

A real-time target detection algorithm is proposed and used in the embedded graphic processing unit (GPU). In view of the lack of computing units and the slow processing speed for an embedded platform, an improved lightweight target detection model is proposed based on the YOLO-V3 (You Only Look Once-Version 3) structure. This model is first trained off-line with vehicle targets and then deployed on the embedded GPU platform to achieve the online prediction. The experimental results show that the processing speed of the proposed method on the embedded GPU platform reaches 23 frame/s for a 640 pixel×480 pixel video.

A stellar image denoising method for active pixel sensor (APS) based on dynamic sequential noise template is proposed. In this method, a dynamic sequential noise template is constructed based on the short-term static background noise and the random noise threshold, which is used to realize the denoising treatment of APS stellar images. The original APS stellar image of the ZY-3 is used as the test data and the research results show that the proposed method can eliminate strip and edge noises very well. Compared with the traditional denoising method, the proposed method can improve the extraction accuracy of the stellar image centroid by approximately 12.98″. Compared with the global threshold segmentation method, the proposed method can improve the extraction accuracy by about 2.16″. Compared with the static background noise method, the proposed method can improve the extraction accuracy by around 9.64″. These fully indicate that the proposed method can effectively improve the positioning accuracy of APS stellar centroid and can be used in attitude post-processing.

To solve the registration problem of a three-dimensional (3D) point cloud under disorder, data occlusion and noise disturbance, a scale point cloud registration algorithm in high-dimensional orthogonal subspace mapping is proposed. The point cloud to be registered is scaled up to complete the affine registration according to the energy-power ratio. The registration accuracy of the proposed algorithm is comparable to that of the classical iterative closest point (ICP)algorithm when the point cloud is out of order with data occluded, size scaled and noise disturbance. Compared with the classical ICP algorithm, the proposed algorithm improves the registration efficiency of the Bunny point cloud data by 98% and the registration speed of the Dragon point cloud data by at least 20 times. Moreover, in the registration of the large-scale Dragon point cloud data, the registration time of the proposed algorithm is 6210.4 s less than that of the classical ICP algorithm, and the registration accuracy is higher than those of other algorithms. The proposed algorithm does not fall into the local minimum and possesses obvious advantages in terms of fast and accurate registration and stability.

Based on the thermochromic phase transition characteristics of VO2, a tunable mid-infrared broadband absorber with a W/VO2 square nano-pillar array is designed. The influences of structural parameters on the absorption performances and the electromagnetic field intensity distributions within the structure, and the absorption characteristics of the absorber under different polarization states and incident angles are analyzed by the finite difference time domain method. The results show that the infrared light incident on the absorber is converted into heat and consumed when VO2 does not undergo the phase transition, and the average absorptivity of the absorber in 3-5 μm wavelength region is as high as 96.2% under the optimal structural parameters. In contrast, the absorber exhibits a strong reflection and inhibits light absorption when the phase of VO2 is changed into a metallic state, and the average absorptivity difference between high and low temperatures can reach 74.1%. The absorptivity of the absorber is less affected by the incident lasers with different polarization states and has wide-angle absorption characteristics, so the absorber is expected to be applied in the field of infrared intelligent optoelectronics.

Fundus retinopathy is the cause of most ophthalmic diseases. Optical coherence tomography (OCT) has been widely used in the diagnosis of ophthalmic diseases because of its non-invasive and rapid imaging features. In this paper, a dither removing method is proposed for three-dimensional OCT retinal image based on curve fitting. The OCT retina image boundary is extracted by preprocessing, and the offset of each frame slice image is calculated by least squares curve fitting. The results show that the proposed method has a significant correction effect on the distortion of the OCT retina three-dimensional reconstructed images.

The rat brain trauma model is established with Feeney's method. The rat brain tissue slices are imaged with the terahertz transmission imaging system, which shows that the rat brain traumatic region has lower transmittance than the normal region. The three-dimensional modeling of rat brain is realized by the three-dimensional reconstruction technique. The model clearly reflects the spatial distribution of the traumatic region inside the rat brain. This study suggests that the three-dimensional reconstruction technique based on terahertz multi-depth slices imaging has great potential in the precise and accurate biomedical diagnosis.

To solve the problem that conventional selective activation methods cannot share the scanning path in laser scanning confocal microscopy and is difficult to be coaxial with the imaging path, we design a coaxial scanning real-time light stimulus system. The field-programmable gate array controls the acousto-optic modulator to modulate the light source of the laser scanning confocal microscope. After the light path scanning, the output of stimulus can be synchronized with the imaging scanning. The system uses the peripheral component interconnect express hard core integrated by the field-programmable gate array in Xilinx KC705 development board, and uses the peripheral component interconnect express interface to transmit the stimulus image of the host computer to the field-programmable gate array. The test results show that the proposed system can satisfy the transmission requirement and adjust the light stimulus area effectively.

By analyzing the propagation and evolution characteristics of a spiral phase modulated orbital angular momentum (OAM) light field, we propose a new approach to investigate the nonlinear effects of the evolutionary wave source of the optical field. Based on the three-wave coupling model and the Collins path integral theory, we obtain the analytical expressions of the frequency-doubled light and its diffraction propagation solutions of the OAM evolutionary wave source. We experimentally study the propagation property of the second harmonic wave by imaging the OAM evolutionary wave source in the frequency-doubling crystal based on a femtosecond Ti: Al2O3 laser and the 4f coherent system. The frequency-doubling output powers of OAM wave sources with different orders are measured and compared with those of OAM modes after evolution. The research results show that using the starting point of evolution as the nonlinear interaction area can solve the problems such as the low nonlinear interaction efficiency caused by light overlapping and the overlarge spot of a high-dimensional OAM beam. Simultaneously, it provides an important basis for the high-efficiency nonlinear manipulation and control of a high-dimensional OAM light field.

According to the requirements of ultra-light and high thermal stability of a space camera, an integrated backplate structure is designed so that the supporting backplate is not only the backplate of the main mirror, but also the main bearing plate of the space camera. The SiC with high specific stiffness and high thermal stability is used as the backplate material. The layout of back ribs is determined by the variable density topology optimization with the addition of a minimum size constraint. The size optimization design is completed by a multi-objective optimization model with Non-dominated Sorting Genetic Algorithm II (NSGA-II), integrating the mirror surface shape and the mass of the backplate. The mass of the backplate is only 0.591 kg and the minimum rib thickness is 2.1 mm. The dynamic and static performances of the optimization results are finally analyzed by the finite element analysis. The results show that the root-mean-square value of the mirror shape in the mirror assembly is 0.158 nm under a temperature rising load of 5 °C, which means good thermal stability. The root-mean-square value of the mirror shape is 1.169 nm and the peak to valley value is 5.403 nm under the X-direction gravity load (the direction perpendicular to the optical axis/the direction of detecting surface shape). The first-order intrinsic frequency of the mirror assembly is 397 Hz and the random vibration response value of the sampling point of the mirror edge is less than 16g. These mean the requirements for space application are satisfied.

An alignment scheme based on the principal component analysis is proposed. The sensitivity matrix of ideal off-axis three-mirror anastigmatic system is processed to obtain weights of system misalignment to residual aberration sensitivity. The reasonable alignment scheme is formulated by removing adjustments with low sensitivities. Based on the principle of computer aided alignment technology of the reverse optimization method, the simulation alignment and the alignment scheme are verified. The results show that with the proposed alignment scheme, the alignment accuracy is up to 10-7 magnitude, and the adjustment amount is reduced from 11 to 6. The alignment difficulty is reduced, and the stability of optical and mechanical structure is improved.

Yellow phosphorescent organic light-emitting device is prepared, and the structure of the device is ITO/HAT-CN/TAPC/TCTA/POAPF…PO-01/Bphen/LiF/Al. The efficiency roll-off of the device fits the triplet-polaron quenching model well. Then, electron-only and hole-only devices are designed. Experimental results show that the holes are majority carriers in light-emitting layer and PO-01 traps holes. The efficiency of the device rolls off drastically due to excitons quenching caused by PO-01 trapping excess holes at high current density. The method of N-doping is used to increase the electron injection, which reduces excess holes in the light-emitting region, improves the carriers balance of the device, and alleviates the excitons quenching caused by excess holes, thereby improving the efficiency roll-off.

The opto-mechanical system design of self-developed irradiance spectroradiometer in reflective solar bands is introduced. The instrument has a spectral range of 400-2500 nm, including three spectral modules, and it can achieve automatic long-term measurement of total solar irradiance, diffuse sky irradiance, direct solar irradiance and the ratio of diffuse irradiance to full irradiance. In order to verify the rationality of the opto-mechanical design, we analyze the field experimental data. Comparing with the ratio data of the traditional manual measurement, it is found that the ratio deviation of the two methods is less than 2%. The instrument has the same irradiance measurement accuracy comparable to the traditional measurement method, and has certain advantages in the automatic calibration of satellite remote sensor.

To satisfy the need of a wide working spectral range for spectral-splitting elements in hyperspectral remote sensing applications, a method for suppressing multi-order transmission peaks of tunable Fabry-Perot filter (TFPF) is proposed. According to the phase conditions of transmission maximum occurrence, the effect of multi-order transmission peaks is eliminated by partition of working spectral range, designation of interference order and determination of cavity length variation region of TFPF. The method can effectively extend the free spectral range (FSR) TFPF and make its spectral scanning properties meet the application requirements of hyperspectral remote sensing.

The uniformity distributions of the anti-reflection (AR) coatings prepared on the quartz tube surfaces by the atomic layer deposition (ALD) technique as well as the influences of the inner diameter, outer diameter, and quartz tube length on the uniformity distributions of the deposited coatings are studied. With the single-wavelength AR coatings as the research object, the reflectivity spectrum from the experimental test matches well with the simulated result. The minimum reflectivity on the quartz tube surface is decreased to 0.17%. If the effect of the sample holder is ignored, the non-uniformity of the AR coatings for the outer and the inner surfaces of the quartz tubes is roughly identical,which is within the range of ±1.69%. The thicknesses and the central wavelengths of the AR coatings for the inner and outer surfaces are almost identical. Moreover, the variation of the quartz tube size has no obvious effect on the coating uniformity of the outer and inner surfaces. Therefore, the AR coatings with small thickness deviations and similar uniformity distributions can be deposited on the outer and inner surfaces of large curvature elements by the ALD technology.

In order to solve the problem that there exists the photothermal effect during the trapping by a metal optical nanotweezer structure, an optical tweezer structure with silicon-based double nanoballs is designed. The three-dimensional finite element method in frequency domain is used to compare and analyze the enhanced field distributions as well as the thermal effects under the same trapping potential energy of polystyrene nanoparticles for the structures with silicon-based double nanoballs and with gold-based double nanoballs. It is found that the silicon-based structure has a relatively low thermal effect and good trapping stability under a high light field intensity. The trapping of polystyrene nanoparticles by the designed silicon-based double nanoballs is simulated, and the trapping forces under a stable field at different positions of nanoparticles with different diameters are studied. The research results show that the silicon-based optical nanotweezer structure can trap nanoparticles stably and simultaneously suppress the structural thermal effects efficiently.

The self-developed portable Raman spectrometer is used to analyze the Raman spectral characteristics of nephrites from Gansu, Qinghai, and Xinjiang, and the spectral differences are analyzed. The nondestructive identification of nephrites from three different origins is realized based on the two pattern recognition algorithms of Mahalanobis distance discriminated method and random forest discriminated method. The results show that the Mahalanobis distance discriminated method and the random forest discriminated method can be used to distinguish the nephrites from different origins with the same Raman peak, and the identification accuracies of them are 87.5% and 95.83%, respectively.

A testing method of fluorescence based soft X-ray absorption spectroscopy is proposed. The method overcomes the problem of low fluorescence yield in soft X-ray region and eliminates self-absorption effect of fluorescence, and the soft X-ray absorption near edge structures (S-XANES) of materials are obtained by partially fluorescence yield (PFY) mode. Through the PFY mode, the soft X-ray absorption near edge structures of the buried element in perovskite solar cells, the trace element of catalyst, and the wide bandgap semiconductor are investigated. Compared with the total electron yield mode (TEY), the developed PFY mode of S-XANES has the advantages in characterization of materials bulk property, insulating materials and samples with low concentration.

An analysis model based on boundary limits for multi-channel spectrograph with the broad band and high throughput is built. The relationships among performance requirements, initial parameters, project budget, and risk of the multi-channel spectrograph are discussed. The mathematic model can get structural parameters of each sub-system of the multi-channel spectrograph quickly according to the given system requirements. It also provide a valuable method to evaluate the feasibility of the design and cost budget reasonably at the beginning of the project. A multi-channel spectrograph based on volume phase holographic gratings is designed for the 4 m telescope. It has a wavelength range from 350 nm to 1000 nm and resolution of 5000 at the blazed wavelength in each channel. The peak efficiency of the whole spectrograph is over 53%, and the monochromatic enclosed energy at 80% is better than 15 μm with the whole working band, which meets the requirements of system performance.

A new method for identification of adulterated sesame oil is proposed, and the method combines three-dimensional fluorescence spectroscopy with wavelet compression and alternating penalty trilinear decomposition (APTLD) algorithms. The three-dimensional fluorescence spectra of pure sesame oil and adulterated sesame oil samples are measured with fluorescence spectrometer, and the error caused by the instrument is eliminated via excitation correction and emission correction, then the true three-dimensional fluorescence spectrum data of the samples are obtained. Wavelet compression is used to compress the processed real data to reduce redundant information. The compression fraction and data recovery fraction are greater than 94.00% and 98.00%, respectively. The APTLD algorithm is used to qualitatively and quantitatively analyze the compressed three-dimensional fluorescence spectral data. The obtained recovery rate is 97.0%-99.8%, and the root mean square error of prediction (RMSEP) is not larger than 0.120. The research results show that the proposed method can accurately identify the pure sesame oil and the adulterated sesame oil samples, and predict the component of the samples.