A postnonlinear unmixing algorithm was presented for hyperspectral images based on backtracking optimization to improve the unmixing accuracy. On the basis of the postnonlinear mixing model, the reconstruction error between the observed images and the reconstructed images was used as the objective function, backtracking search optimization algorithm was used to search in the solution space to obtain the optimal solution which minimize the objective function. In the search process, the boundary control mechanism of the backtracking search optimization algorithm effectively ensured the constraint condition in the hyperspectral image unmixing, and then the abundance and nonlinear parameters can be estimated accurately. The experiments conducted for both synthetic images and real remote sensing images show that the algorithm proposed is provided with excellent unmixing performance. The unmixing accuracy achieved is significantly better than the state-of-the-art nonlinear hyperspectral images unmixing algorithms.(15JCYBJC17100)

In order to relieve the coupling effect of 4 m lightweight mirror support system, a generalized least square method was proposed to calibrate active optics. Firstly, the support system of 4 m lightweight mirror was introduced, and the constraints needed to meet in hydraulic Whiffletree passive support system was deduced. Then the Zernike polynomial corrected by node area weighted factor was used to fit the mirror deformation, resulting to influence matrix between Zernike coefficients of mirror deformation and active forces. After that, a generalized least squares method was adopted to obtain the optimal correction forces which simultaneously satisfied the equality and inequality constraints. Finally, the proposed method was applied to the calculation of the optimal active forces to correct the mirror deformation caused by the gravity, and the influence of different damping factors on the calculation results were analyzed. Results show that the optimal correction effect is achieved when the damping factor is 7.4e-9, and the RMS of the mirror deformation is changed from the original 271.5 nm to 8.3 nm after correction, verifying the feasibility of the generalized least squares method to the calibration of the 4 m lightweight mirror active optics.

During the quantity production of small size photoelectric encoders, not only the encoder dynamic error must be detected in factory inspection, the error of non-standard encoder must also be traced and corrected. In the process of the dynamic error detection of photoelectric encoder in high speed and low speed, the reason caused excessive dynamic error of photoelectric encoder must be quickly found out, so that the encoder can be adjusted according to the reason of excessive error. The measurement and estimation system for dynamic errors of photoelectric encoder was designed by proposing a measurement and estimation method. Firstly, by the analysis on dynamic error of photoelectric encoder, the theory of dynamic error component was proposed. Secondly, the estimation method of dynamic error based on AR model spectrum estimation was proposed, and error criterions were proposed by error estimation. Thirdly, the measurement and estimation system for dynamic errors of photoelectric encoder was designed, error detection and estimation were achieved by this system. The speed range of this system is 0.5-8 r/s, measurement precision is 2″; the system can clearly display the mean square of errors in every frequency, this result can offer evidence for producer to reduce the error of encoder. The system was exact and reliable, and provides straight forward and effective measurement method for producer.

Several different integral times at multiple temperatures added by the attenuators are often required for the wide dynamic range infrared radiometric systems. It′s not only inefficient, time-consuming and costly, but also affects real-time and mobility when used for the missions. Based on the calibration model considered integral time, the high-speed calibration algorithm was proposed for wide dynamic range and random integral time. The algorithm can obtain calibration equation for wide dynamic range and random integral time mainly by only turning one temperature point and one integral time point. The verifying experiments were performed by the Φ400 mm infrared radiometric systems, for purpose of the accuracy of calibrations and measurements by the algorithm mentioned above. Experimental results show that the high-speed calibration algorithm compared with the direct calibration results maximum error is 2.75%, the maximum error radiant measurement is 11.69% and the maximum error compared to the conventional radiant measurement result is not more than 1.52%. The high-speed calibration algorithm can improve the efficiency of radiometric calibration, shorten the calibration time and achieve the equation of calibration at random time on the premise of the accuracy of consequences. The entire process of mission is real-time, simple operation, high reliability, easy to implement and suitable for the measurement requirements to infrared radiometric system, that can be directly applied to the target range of infrared radiometric missions.

The local adhesion in the X-Ray images of SMT material plates has negative effects on the accuracy of counting and detection for components. A method based on the constraints of position information was proposed, which was used to improve the accuracy of segmentation and detection with the position information between components. Firstly, the center point and the starting components arranged in the inner ring were fit out on the basis of the components in a spiral arrangement rule. Then the dual constraints of normal position and priori position, which were based on the constraint model between the center point and different components, were finished to limit the region of target device. Finally, the segmentation for conglutination between components was completed by dividing the boundaries between components. Experimental results show that this method can improve the accuracy of the result of segmentation and detection. The detection error rate tested on different specifications material plates is controlled within 0.15% under the experimental conditions of 9 216 pixels effective image element and detail resolution of 110 1p/cm.

The principle of measuring nighttime total atmospheric transmittance was analyzed and the equipment was developed to measure weak radiation atmospheric transmittance within any sky zone of elevation angle greater than 15° from level 0 to level 5 magnitude stars during the spectral band of 400 -750 nm, and the bandwidth of measurement was 30 nm. The method of automatically controlling CCD exposure time was used to obtain a sub-band weak radiation signal of stars, using image feedback mode to correct errors in the telescope searching-stars, which realized automatic unattended continuous observation all night. The equipment calibration adopted the combination of Langley - plot calibration method and multi-objective star calibration method. The measurement results were compared with the simulated values of CART and Modtran, which verified the feasibility of the measurement method and the reliability of the equipment.

According to the ray tracing method of the vector field and the geometrical relation of the optical path, the relations between normalized incident height, refractive index and divergence angle have been deduced. With the aim of minimizing the divergence angle, the optimum design parameters of cat′s eye retroreflector have been obtained, and optical design of cat′s retroreflector have been performed by ZEMAX. The polarization ray tracing and polarization properties analysis of cat′s eye retroreflector were carried out based on the three-dimensional Jones matrix method, and the retardance and diattenuation of cat′s eye retroreflector in the exit pupil of cat′s eye retroreflector and the influences on the output polarization state were analyed. The calculation results had shown that the maximum values of retardance and diattenuation were 0.3° and 0.027, which have no effect on the output polarization state. The FRED software was used to simulate the situation of cat′s eye retroreflector with input horizontal, 45° linear and circular, elliptical polarization states, to verify the correctness of previous theoretical analysis. The simulation results have indicated that the output polarization state is the same as that of the input polarization state through cat′s eye retroreflector, that is to say, cat′s eye retroreflector has the polarization state-preserving effect, which is significant for the expand application for cat′s eye retroreflector.

A filter model with one-dimensional photonic crystal structure (AB)N was proposed. Layer A was gallium arsenide (GaAs) material and layer B was an artificial periodic structure material. B was anisotropic and made of aluminum-doped zinc oxide and zinc oxide (AZO/ZnO) alternately. Based on the theory of electromagnetic wave transmission matrix, the transmission rate of photonic crystal was derived. The numerical simulation shows that the center wavelength of the transmission is 1.55 ?滋m and corresponds to the photon pass band. The number of transmission peaks was determined by the period N of the photonic crystal. Increasing the filling factor h of the B layer from 2/3 to 11/12, the transmission peak wavelength was blue shifted and the moving range exceeds 200 nm. Increasing the thickness of the A and B layers, the peak wavelength was red shifted. When the incidence angle increases, the peak wavelength was blue shifted. The photon crystal will maintain a high transmittance within the control range of each parameter. These phenomena provide a theoretical reference for the design of multi-channel tunable high performance filter for optical communication.

A novel scheme of optical phased array(OPA) based on wave-guide was represented in this paper. Fiber paths was main design of system, the single mode fibers were used as transmission paths, photonic crystal fibers(PCF) were adopted as the output array, LiNbO3 wave-guide was used as the phase modulator. The system configuration was given, performance of main device such as LiNbO3 wave-guide and PCF array were analyzed. According to the theory of OPA and electro-optical effect of LiNbO3 wave-guide, the feasibility of system had been demonstrated. By adjusting the phase shift of each LiNbO3 wave-guide, the beam deflection had been observed. Simulation experiments had been implemented to study the influence of its structure parameter on output diffraction characteristics. The results show that the inter-elements distance, the quantity of fiber core and arrangement of fiber core affect the beam scanning quality including Full Width at Half-Maximum(FWHM), output intensity distribution.

Light propagation characteristic of dual defect microcavity inserted or replaced by photonic crystal were studied by numerical calculation and simulation. The result shows that when the periodicity of dual defect becomes larger, there will be more defect modes in microcavity transmission spectrum, the modes in replaced dual defect microcavity are more than that in inserted dual defect microcavity. As the high-refractive medium of defect thickens, the defect modes in dual microcavity draw close to the center and tend to degenerate; at the same time, pairs of double defect modes appear on both sides of band gap, with more on the high frequency side and less on the low frequency side; however, the defect mode quantity in replaced dual defect microcavity is larger than that in inserted dual defect microcavity. While the low-refractive medium of defect thickens, defect modes of inserted dual defect microcavity move to low frequency and tend to separate for function coupling, with the transmission index decreasing; but the defect modes of replaced dual defect microcavity draw close to the center and tend to degenerate, and meanwhile the whole defect modes slowly shift to high frequency and the transmissivity remains 100%. Theses light propagation characteristic provide design reference for optical filter, switch and laser devices.

As one of the most advanced optical detection techniques, photoacoustic spectroscopy has been successfully used in the area of trace gas detection. Realizing a multi-gas photoacoustic spectroscopy sensor while keeping a high detection sensitivity was the design purpose. For the detection application, a multi-gas photoacoustic spectroscopy sensor based on the Helmholtz resonance was designed and optimized. The sensor was modeled by using the finite element method and optimized for its parameters, including the dimensions of excitation cavity, connecting tube and detection cavity. Also the temperature and pressure properties of the sensor were investigated. As a result, the detection sensitivity could be improved by optimizing the parameters and controlling the temperature and pressure. The optimizing result shows that the gas sensor has a best performance with the infrared light source IR-19, and parameters with excitation cavity of Φ38 mm×9 mm, connecting tube of Φ5.9 mm×10.2 mm and detection cavity of Φ30 mm×5.8 mm. Tested by the experiment, the detection limit of 5.08 ppm is achieved for the CO gas detection. The study results also provide the reference for design of photoacoustic sensor in multi-gas detection.

The bottle beam was analyzed by using the discrete Fourier method for the first time in this paper. The bottle beam intensity distribution was described by discrete Fourier transform, then Sellmeier formula was added to adapt to a LED light source with some spectral width. By considering sampling and discrete variables, discrete Fourier's formula of bottle beam generated by a LED light source was derived. At last, the parameters were substituted and the intensity distribution of the beam at different distance was obtained by the numerical simulation with MATLAB. Experiments were designed to verify it, and the experimental results showed that using LED as light source can produce high-quality bottle beam. Numerical simulation is in line with experimental results basically, which proves the discrete Fourier transform method is feasible to analyze bottle beam generation by incoherent light. Furthermore, compared to conventional scalar diffraction theory, it is more simple and fast. It′s a new and reliable way to study bottle beam.

Signal distortion caused by High Peak to Average Power Ratio(PAPR) is the major disadvantages that the Free Space Optical(FSO) Orthogonal Frequency Division Multiplexing(OFDM) system has to deal with, while Selected Mapping algorithm (SLM) can effectively solve this problem. The research on PAPR, however, is mainly focused on theory simulation, and especially on the reduction of PAPR at the transmitter. Therefore, works on the implementation of the algorithm on PAPR reduction in FSO-OFDM system, especially the verified way of reducing PAPR at the receiver, are of significance in literature and practice. It was verified that the performance of FSO-OFDM system on noise immunity can be increased after using the SLM algorithm through simulation, and effective when applied to the channel transmission in hardware experiments. And this experimental results lay the foundation for follow-up studies on the reduction of the PAPR.

Bit synchronization is one of the key technologies of space laser communication and measurement and control unified system. Its accuracy directly affects the performance of the whole system. Bit synchronous technique based on Gardner algorithm is easy to be interfered by self-noise of system. A self-noise abatement Gardner algorithm was proposed and a synchronous loop was presented based on this algorithm. Theoretical analysis and simulation results show that the self-noise abatement Gardner algorithm and the synchronous loop can effectively reduce the timing jitter caused by systems self-noise and have a faster bit synchronization establishing time when Eb/N0 is greater than or equal to 9 dB.

Aiming at the combined effects of geometric spreading and pointing errors on space optical links, the bit error rate (BER) performance of the all-optical amplify-and-forward(AOAF) relaying high altitude platform(HAP)-GEO-HAP communication links was researched. The GEO-based satellite AOAF relaying HAP optical communication link was modeled. The statistical characteristics of the channel fading index and the strength of the noises were investigated when the background radiation and the amplified spontaneous emission(ASE) noises were considered. With the help of the obtained Q factor of the proposed link model, the expression of BER was theoretically derived. The relationship between the BER performance and the bit rate under different parameters such as the transmitted optical power, jitter standard deviation and the beam-width was analyzed by simulation. The results show that, the BER performance is obviously improved by increasing the transmitted optical power, reducing the jitter standard deviation or decreasing the beam-width when the bit rate is low. On the other hand, a slight improvement of the BER performance is obtained with the same measures when the bit rate is larger than 4 Gbps.

The three-dimension thermal distributions in double-cladding high-power Thulium-doped fiber amplifier(TDFA) with different inner-cladding shapes were achieved by solving the analytical thermal conductive equation with different inner-cladding boundary conditions. It indicates that the temperature difference induced by the overlap factor of double-cladding fibers(DCFs) with different inner-cladding shapes can be up to 107 K in the core. Moreover, the distance between the splice point and the position with maximum temperature relies on the ratio of the seed power to pump power and it can be 30 cm when pumped with 100 W and seeded with 10 mW. By analyzing the transverse and longitudinal thermal distribution, it is demonstrated that the offset DCF may be a better choice for TDFA due to its lower maximum temperature, high pump efficiency and Gaussian thermal distribution on the cross-section.

Optical vortex(OV) has been researched and applied in various front fields. Here, a novel method, by combining vortex retarders and half waveplates, was proposed to generate OVs with arbitrary topological charges. As the optical transmissivity of the waveplates were high, the energy efficiency of OV with topological charge of 3 was still higher than 93% in experiments. Further, the interference method was employed to detect topological charges of the generated OVs, by the direction and number of forks in the fork like pattern. As last, surface plasmon polariton (SPP) field was excited by highly focused OV with topological charge of 3, and gold particles were attracted and manipulated in the plasmonic vortex field. The results demonstrate significance of such method, and it will express enormous potential in various applications.

Instrument transfer function(ITF) of white light interferometer(WLI) influences the measurement accuracy of power spectral density(PSD) for optical surfaces. To understand the nature of ITF, WLI was treated as an incoherent optical system and a sinusoidal surface was the input signal. Studying the Bessel function expansion of interference intensity spectrum of sinusoidal surface, the theoretical analysis indicated that the low pass of optical system led to the surface height attenuation. The nonlinearity of Bessel function resulted in the nonlinear response of WLI to height, and the response approached linear when the height was much less than wavelength. Numerical simulations and experiments validated the theoretical conclusions. Step-artefacts of 30 nm, 80 nm and 120 nm height were employed to calibrate the ITF of a commercial WLI. The experimental result and the simulation result match each other well. ITF rises when the surface height increases, which manifests obvious nonlinearity. If the surface height is less than λ/10, ITF nearly equals the modulation transfer function of WLI optical system, in which case WLI is approximately linear to surface height. This paper has promotion for understanding WLI transfer characteristic and enhancing PSD measurement accuracy.

In order to develop related researches of infared window, it is necessary to simulate surface temperature response of infrared window exactly. As the property of infrared transparence, it is hard to obtain the surface temperature of ZnS without disturbing the flow field, thus it is not possible to simulate the temperature response process of infrared window. A method was introduced to simulate the surface temperature response of ZnS infared window by arc-heated wind tunnel. First, the surface temperature of ZnS was measured by using 2Cr13 which had similar thermophysical properties with ZnS. Then, the temperature response process of infrared window was simulated. Finally, several tests were taken to identify this method by arc-heated wind tunnel. The result shows that 2Cr13 and ZnS have the same temperature response properties when tested in arc-heated wind tunnel flow field at temperatures between 150-250 ℃ . It is reliable to measure the surface temperature of ZnS by using 2Cr13 which has similar thermophysical properties with ZnS in arc-heated wind tunnel simulation test.

A infrared single-photon detecting system has greatly improved the performance of the weak light detection system due to its ultra high sensitivity and SNR. Based on TSMC 0.35 μm CMOS process, a design method for digital mode infrared ROIC and its application system were proposed. The system included high-segment pixel TDC and low-segment global shared TDC controlled by OSC multiphase clocks. The implemented circuit was operated at frame rates of 1 kHz with data output in a series mode, also it has a time resolution of less than 1ns along with a 3 μs measurement range under 170 MHz reference frequency condition, which can finally satisfy the application of ranging and imaging after relevant data processing.

Radiators used to maintain space target functioning normal were introduced. Operating mode and working condition were presented. Current used space target infrared characteristics models were summarized. With further analysis, the outside surface of space target were divided into common area and radiator, and energy equations were built separately. Taking FY-1C as an example, and taking account of orbit, materials and structure, temperature field of the outer surface of space target was calculated using finite element method. When dissipation power was 0 W and 100 W, the biggest temperature difference of radiator was 51.49 ℃. Analyzing the temperature field with orbit, the illumination in entrance pupil of detection system in a distance of 5 km was calculated. When space target was in the shadow of earth, during which the radiation from earth or solar radiation reflected by earth was negligible because of the big incident angle, the illumination of the space target differed by 1-2 orders of magnitude. When space target was under the sun, different dissipation powers can only affect the illumination of long infrared spectra apparently, because of the reflected radiation of target.

Research on scattering characteristics of sand and dust in the infrared waveband has important theoretical significance for Free Space Optical(FSO) communication. According to some infrared wavelength commonly used in FSO, its scattering characteristics in different particle sizes of sand and dust were analyzed by Mie theory; based on log-normal distribution model of particle size, the attenuation coefficient of laser in waveband 0.76-10.6 μm were calculated under the condition of sand and dust. The results show that: the transmission attenuation near waveband 7.4-8.0 μm are minimum under the condition of sand and dust, and it′s significantly lower than that in other wavebands. Therefore, the laser in waveband 7.4-8.0 μm should be chosen to be used in FSO under the condition of sand and dust, for that its transmission performance will be obviously better than that in other wavebands. At last, take wavelength 1.064 μm as an example, the "visibility" with the change of particle concentration was analyzed, and the result was compared with that by Monte Carlo method.

An algorithm of infrared vehicle detection under the complex background was propose. First, a novel self-adaption piecewise linear stretching function was utilized to enhance the targets when the overall intensity of image was low. Second, the saliency map of the enhanced image was used to generate the ROIs of the potential targets. Then, the vehicle targets were able to be detected by the edge-based segmentation in ROIs using average gradient. Finally, multiple features were fused to discriminate the belief of a region belonging to a vehicle target. Experimental results on the real infrared images show that the algorithm can effectively detect the vehicle target on ground.

In order to study the effect of light intensity uniformity on the conversion efficiency of GaAs cells, based on the working principle of single junction GaAs cell, the photoelectric conversion efficiency of GaAs cells was analyzed when the laser intensity was different by using the equivalent circuit, and the conversion efficiency of GaAs cells in different light intensity uniformity was studied by experiments. Results show that the light intensity uniformity has a great influence on the photoelectric conversion efficiency of the single junction GaAs cell. In some extreme conditions, it may cause the hot spot effect and completely destroy the GaAs cells.

ZrO2 ceramic coating was designed for aluminum alloy plates and their anti-laser performance was tested using 976 nm CW laser. The alloy substrate is 2.5 mm thick and the coating is 0.3 mm thick. In the tests the plates were subjected to 0.3 Ma tangential airflow. Temperature of the rear surface was recorded with thermocouples. Spectral reflectance of unirradiated area and irradiated area was measured respectively, and XPS analysis was done to investigate the change of elemental composition in the coating induced by laser irradiation. The experimental results show that, the sample could withstand 60 s irradiation by 976 nm CW laser with power density up to 700 W/cm2, and that the color of irradiated area turns white from offwhite, and that the reflectance to near-infrared light increases. The change in color may be the result of pyrolysis or gasification of carbon-containing compounds during laser irradiation, which is consistent with the results of XPS analysis.tangential flow

The effects of varied laser conditioning parameters on the damage morphology change process of HfO2/SiO2 reflective film under nanosecond pulse were experimentally studied in this paper. By comparing of damage expansion and the influence on spectra for two types of damage, the role of laser conditioning is illuminated to restrain the process of damage morphology change process. The two contrary impacts of laser conditioning on thin film, nodule clean up and absorbing precursor formation, were revealed by damage morphology analysis. The experiment result shows that the contours of damage morphology change thresholds will shift to higher laser fluence and more shots after laser conditioning. In addition, considering efficiency and profit of laser conditioning, the combination of higher laser fluence and two scanning steps could have the best choice in morphology changing elimination, and the morphology change threshold could increase to 140% in the best case.

The influence of laser resonator misalignment on laser performance was investigated theoretically and experimentally. In theory, based on the character of quasi-continuous diode-pumped electro-optical Q-switched solid-state laser, the rate equation model including the modifying function of cavity misalignment was set up. The influence of cavity misalignment on laser losses and laser performance was simulated with different laser modes, laser beam radius, radius of iris and position of iris and the influence of cavity misalignment on laser performance was simulated by model′s numerical simulation. On the other hand, the influence of resonator misalignment on laser performance was measured in experiment. For the concave-convex cavity with cavity length of 250 mm and iris radius of 3.5 mm, the influence of concave mirror′s misalignment on laser performance was more serious than that of the convex mirror′s misalignment. The energy decreased from 220 mJ to 150.9 mJ corresponding to decreasing of 31.4% and pulse width increased from 5 ns to 6.5 ns with the concave mirror′s misalignment angle increasing from 0 to 490 μrad. The experimental results and theoretical results are in good agreement, indicating the validity of the laser model. The established laser model and the utilized experimental measuring method can be effective means to obtain the influence of resonator′s misalignment on laser performance, which can be a reference of the design of high reliability laser.

Ring laser gyro is usually pumped by high voltage discharge. Negative impedance characteristic of discharge tube can cause instability or oscillation of the current adjustment circuit. In order to optimize the discharge circuit loop, the impedance characteristic of discharge tube in ring laser gyro was researched. The voltage drop was measured under different discharge current, and voltage-current characteristic was obtained through nonlinear curve fitting. And then differential impedance, which was related to adjustment stability, was analyzed. The research can provide guidance for selection of ballast resistor and reduce the power consumption on premise of reliability.

In order to produce laser projectors in large quantities, a green laser with compact structure stable performance and low cost as the green light source was need. In this paper, a compact microchip laser array was obtained, utilizing optical-contact Nd:YVO4/PPMgOLN bar as gain media and frequency doubler, combined with soldering packaging technology. Three green beams of the laser were obtained with the total output power of 223.7 mW, each laser beam had good spot profile and stability, the microchip laser array had compact size of 47 mm×35 mm×25 mm. It had a fluctuation of less than ±2.5% for 2.5 h. Various performances of the laser would fully meet the requirements of laser projector as the green light source, and it could be mass-produced in low cost.

A Laser Diode(LD) end-pumped Nd:YLF quasi-continuous 351 nm ultraviolet laser with high efficiency and high peak power was demonstrated under acousto-optical Q-switched operation, by using extracavity frequency-doubled and frequency-tripled with two LiB3O5(LBO) crystals cling to the output mirror. An average output power at 351 nm of 450 mW was obtained at repetition rate of 1 kHz, pumped power of 14 W and output power of fundamental wave of 1.45 W. The optical-to-optical conversion efficiency is up to 31.04% and pulse width is 7.5 ns corresponding to the peak power as high as 60 kW.The beam quality is satisfactory.工系统方面的研究。

Stable volume discharge of gas mixture is the foundation for discharge pumped gas laser, and preionization is one of effective solutions to stable volume discharge. The design of preionization pins of discharge-initiated pulsed HF laser was introduced, and preionization effects were evaluated by numerical simulation and experimental investigation. The simulation results show that the preionization electron density is about 109/cm3 in discharge area. The experimental results show that the output energy of laser with preionization is several times higher than that of without preionization in lower working voltage for SF6-H2 mixture, and for SF6-C2H6 mixture, the output energy of laser increases from 200 mJ to 297 mJ with preionization and the charging voltage of 20 kV, which is about 50% higher than that of without preionization. The simulation and experimental results show that preionization has effects on controlling and maintaining volume discharge and improving the uniformity of discharge.

The non-uniformity correction algorithm of infrared image using Lagrange interpolation algorithm to calculate the offset coefficient was proposed in this paper. It overcame the shortcoming of periodic calibration for the calibration algorithm and it was less complex than other algorithms based on the scene. According to the characteristics of large amount calculation, the method of constructing SOPC system based on FPGA was used to realize the algorithm, including parameter calculation, background acquisition and calibration model. Parameters were calculated and updated according to a certain strategy. Background information module was used for collecting the information of the uniform background. Correction module was used to complete the real-time non-uniformity correction. The experimental results show that this system has the advantages of less resources, shorter delay and better imaging effect.

It′s significant to research the infrared radiation model and simulate infrared image of unmanned aerial vehicle(UAV). The interpolation calculation of skin temperature was realized, based on the database of steady-state temperature distribution of the UAV which was calculated in use of CFD. Due to the new method of grid number projection, the orthogonal projection of depth buffer was improved and a two-dimensional projection algorithm for infrared image was implemented. The IR model of UAV was finally built to calculate spectral and band of infrared image, in which atmospheric transmission and detector response were both considered. The UAV infrared characteristics of the whole trajectory and calculation speed of this model were analyzed. The simulation value was compared with the measured flight experiment data, the results show that the relative error is less than 20% and can be used for experiment test and analysis.

Laser protection is a technology of wide application. Aimed at the problem that strong absorption in visible wavelengths and equipment or operator injury caused by laser high-reflective film specular reflection exist in infrared laser protection technology, an infrared laser non-specular reflection optical micro-structure formed on optical material surface was presented. It had little effect on visible light transmission and large-angle scattering to 1 064 nm laser. Light track method was used to design double-side micro-cylindrical lens arrays with a dislocation construction. Array period T and curvature radius of lens units R should meet the condition: 0＜T＜R/2■. Virtual-Lab optical modeling software was applied for the simulation of designed micro-cylindrical lens arrays. The results is that, average transmittance rate of visible light drops 7%, which has little impact on practical result, and which can be made up by visible wavelengths fabrication antireflection coating; 1 064 nm infrared laser reflection is greater than 75%, divergence angle is greater than 30°. Fabrication experiment of micro-cylindrical lens arrays was finished by digital mask lithography technology, the test result was similar to simulation result. The conclusion is that, micro-cylindrical lens arrays can cause wide-angle scattering, which greatly reduces the single-directivity reflection echo energy of laser to achieve the purpose of laser protection.

The biological material has the characteristics of rich morphology, wide particle size distribution, short growth cycle, low mass density. It is significant to study its performance characters for development of new extinction material. In order to study the diffusion characteristic of biological extinction material, according to Rechtman diffusion theory, the diffusion model of biological material was established. The diffusion of biological material in typical natural environment was simulated, and the effects of wind speed and surface roughness on the diffusion of biological material were quantitatively discussed. The results indicate that the maximum value of concentration of biological material on reference point was continuously reduced, and the spatial position of maximum value maximum value of concentration was also changed. The higher the wind speed and the greater the surface roughness, the faster the diffusion rate of biological material was.

The large-aperture bionic compound eye optical system can have a large field of view, it overcomes the defects that the aperture of biological compound eyes is small while the vision distance is short. The relationship between the actual array period and theoretical calculated period number was researched through analyzing the field stitching method of bionic compound eye, the method called "padding sub eye method" was proposed to reduce sub eye configuration density. By studying the geometry relationship of the fixed spherical body design, the mathematical connection between the main body vertical radius d and the curvature R was established, the design method of fixed spherical body was proposed, and the proposed bionic compound eye field stitching theory was perfected. Finally, the correctness of the design method through the actual alignment experiments was verified. This method can improve the integration of the bionic compound eye system, reduce system size, and promote the practical application of bionic compound eye optical system.