A freeform off-axis three-mirror system based on a freeform mirror was designed for the long focal length and wide-field off-axis three-mirror system in a space telescope. To obtain a system with a focal length of 4 500 mm, field of view of 11° and the ratio of the total length of system and the focal length in 1/3, the design and performance between traditional off-axis three-mirror system and freeform off-axis three-mirror system were compared. In the designed optical system, the freeform mirror was introduced to the secondary mirror to enhance system optimization and improve the balance capacity for optical aberration. After the comparison and optimization, the relative aperture of 1/9.5 was chosen.In the system, the average of wavefront error of all field of view is better than 0.030λ(λ=632.8 nm), and the average of the Modulation Transform Function(MTF) of all field of view is greater than 0.434(71.4 lp/mm) that is close to the diffraction limitation. Furthermore, the MTF has im-proved more than 5% compared to that of traditional TMA system and the departure of the secondary freeform mirror and theoretical spherical surface is only 1.1λ(λ=632.8 nm). The highly precision testing can be accomplished only by a standard sphere combined with the non-null testing method based on digital mask, which solves the problem of large-aperture convex freeform mirror testing and is helpful for the manufacturing of freeform secondary mirror. The off-axis three-mirror system based on freeform secondary mirror shows its advantages in smaller cubage, better realizability in engineering, better wavefront error and higher MTF.

A polarized spectroscopy was designed and manufactured based on the laser-produced plasma experiments to diagnose the X-ray polarization of a laser-produced plasma in 0.2-20 nm. The experiment was performed at the 20 J laser facility in Research Center of Laser Fusion, China Academy of Engineering Physics (CAEP). In the experiment, Poly Ethylene Terephthalate(PET) crystals were used as dispersive elements in the two orthogonal directions and the imaging plates with the effective area of 30 mm×80 mm were taken as detectors. The optical paths of the X-ray beam were about 980 mm and 310 mm from the source to the detectors via the PET crystals, respectively. The X-ray spectrum emitted from the aluminum plasmas was recorded by the image plates. It shows that the polarization of the two Al Ⅻ spectra are about 0.123 3 and 0.148 1 after correcting helium resonance and combination spectrum. The experimental results demonstrate that there is anisotropic phenomenon in the laser-produced plasma, and the proposed method is effective to diagnosis of the polarized spectrum of high-temperature plasma.

The calibration of an imaging spectrometer by using the diffuser method should consider not only the accuracy of calibration instruments, but also the effects of instrument and geometric parameters from the imaging spectrometer on calibration results. According to the calibration principle of imaging spectrometer using the diffuser method, this paper deduces the expression of the collected signal electron number in each spectral channel of the imaging spectrometer by taking the parameters as self-variables, and it analyzes the effects of the incident angle of standard lamp, the distance of standard lamp to the diffuser, the slit direction and the observation incident angle of the imaging spectrometer on the electron number collected by the detector pixels. The results show that when the F number is set to be 3 and 4, the entrance pupil in 0.1-0.2 m, the standard lamp illuminating angle less than 12° and the distance of the standard lamp in 0.5-2 m, the change of the signal electron number can be less than 1%.It satisfies the accuracy need of the calibration by using the diffuser method.

The key design of off-axial Three-mirror Anastigmat(TMA) systems was summarized, and an improved design of the off-axial TMA optical systems was proposed with a new optical structure. By taking two designed systems for examples, the optical elements of the TMA system were reasonably arranged, the secondary mirror was designed as a spherical surface, and the primary and the third mirrors were optimized with high-order aspheric coefficients. Obtained results show that the Modulation Transform Function(MTF) curve of the designed system approaches to the optical diffraction limit with an improved value of 0.6 at the spatial frequency of 50 pl/mm. Compared with the common off-axial TMA optical system, the Strehl ratio can be increased from 0.91 to 0.93. The surface tolerance is relaxed from λ/50 to λ/40, and the assembling tolerances of the primary mirror, second mirror and the third mirror can be relaxed by a factor of 4. The improved design makes the manufacture and assembly of the off-axial TMA optical system much easier, and helps the system to achieve excellent performance in the level of optical diffraction limit. These are beneficial to the popularization of off-axial TMA optical systems.

By using an optical fiber coupling structure, a new particle measuring method through introducing Lorentz classic dispersion formula and multi-wavelength processing was proposed to amend the scattering influence of a sample cell wall and the change of the refractive index. Firstly, the theory of total light scattering for the particle measurement was introduced, and the problems existied in the sample cell and measurement method were studied. Then, the Lorentz dispersion formula was introduced to process the refractive index changes caused by dispersion and the multi-wave length processing was taken to obtain the average particle size and particle size distribution. By establishing an object function, a method based on the genetic algorithm was proposed to implement the inversion of particle refractive index. A new sample cell was designed and manufactured with the end face coupling of optical fiber, and the particle sample dripping placed the end face between two optical fibers was scanned by a spectrometer in multi-wavelengths.Finally, through the algorithm of the computer, the average particle size and particle size distribution were measured. The simulation and experiment result indicates that the average particle size and particle size distribution can be well determined by this method in the case of unknown refractive index.Furthermore, this method has advantages of reliability and stability. When a 2% random noise was put in the transmission extinction measurement values at two wavelengths, the inversion errors of average particle size and refraction index are less than 1% and that of the particle size distribution is less than 5%. It can meet the needs of national standard for mean errors and repeatability errors.

In consideration of stray light sources of a star photoelectric-detection system, this paper analyzes the generation of the stray light and its transmission characteristics, and proposes how to evaluate the effects of stray light on the photoelectric-detection system. A reasonable suppression scheme for the stray light is provided for guiding the design of opto-mechanical structure of the system. Then the main baffle, baffle vane and the coating technology are used to eliminate the effect of stray light on the opto-mechanical structure. The opto-mechanical structure of the system is simulated by a software and the results show that their Point Source Transmittances (PSTs) before and after eliminating the stray light have been reduced by two orders of magnitude between 10-8 and 10-10. Meanwhile, the obtained detection data in experimental field demonstrate that the scheme can extract a detection im-age with 5 star magnitudes, and the stray light has been suppressed effectively.

This paper researches the calculation method for the change of space intersection angle of a mapping camera, for the change of the angle should be within 5″ to ensure the reliability of transmitting data in orbit. First, the temperature field was calculated for high and low temperature operating conditions with the I-deas software. Then, the calculated temperature field was input to a structural model as temperature load to complete the temperature mapping from a temperature model to the structural model, namely, from a simple model to a complex model. The thermal-elastic deformation was calculated and the center of optical element deformation was obtained. Finally, the change of space intersection angle of the mapping camera was calculated in two operating conditions with a self-programming software. The results show that the change of intersection angle is below 5″ when the temperature operating condition is changed. It concludes that the calculated method can get the accurate change of intersection angle and the calculated data can benefit to the analysis on the space mapping.

A spectral calibration facility consisting of a monochromator and a collimator was established to calibrate the Offner imaging spectrometer with a convex grating designed by ourselves. The self-designed software for sampling and processing spectral data was used to calibrate the Offner imaging spectrometer and analyze obtained data. Calibrated results show that the calibration facility has a compact conformation, better commonality and high calibration precision, and its spectral range is better than 400-800 nm and spectral resolution is superior to 10 nm, which can satisfy the calibration need in practice.

The primary mirror of a two-reflective system for a telescopy was replaced by a Golay 3 sparse aperture to reduce its weight and volume while remaining high angular resolution. Firstly, the two-reflective system with a spherical primary mirror and a hyperbolic secondary mirror was designed according to the third-aberration theory. Then, the primary mirror was substituted by a Golay 3 sparse aperture, and the relationship between fill factor and sub-mirror radius was analyzed. The whole system was simulated by Zemax software to obtain the Modulation Transform Function(MTF) and the spot diagram of the system . Finally, two silicon correcting lenses were inserted behind the secondary mirror to optimize the radius of the correcting lens and imaging distances and to improve the field of the system. By Zemax simulation, it is shown that the cutoff frequency of the system has been decreased from 114.5 lp/mm to 97.7 lp/mm and to 77.8 lp/mm as the fill factor reduces from 22.2% to 15% and to 10%, respectively. Moreover, inserting the correcting lens in front of the image plane can greatly enlarge the field of the system, which shows the RMS value of maximum spot radius without correcting lens is about 2.5 times larger than that with a correcting lens in the field of 0.3°.

Solar-blind UV induced Transmission Filters (SBU-ITFs) were fabricated by ion-beam sputtering. Firstly, SBU-ITFs were designed and analyzed theoretically, and the deposition of ZrO2, SiO2 and Al films by ion beam sputtering was optimized. The optical constants of ZrO2 and SiO2 films were derived from the transmitting and reflecting spectra, and the optical constant of Al film was determined by the inversion of the variable incidence ellipsometric spectrum of a sample JGS1/SiO2/Al/SiO2/air(SAS).On the basis of the research, the SBU-ITFs were fabricated by using the three films above and their optical properties were measured. The measurement for SBU-ITFs shows that the peak transmittance of the filter at the 264 nm center wavelength is larger than 60%, the transmittance band width (FWHM) is about 13 nm, and the optical density in the 300-350 nm cutoff spectral region reaches to 2.6.The analysis of the transmittance shows that the main factors affecting the properties of SBU-ITFs are a slower deposition speed for Al film and smaller amount of O2 contents remaining in the chamber during the deposition of Al film.

In a fiber strain sensing system, strain sensing cable is usually used as a sensing element instead of bare fiber.Therefore, it is necessary to explore the high stress transfer performance of a strain sensing cable.Combining Lame formula and generalized Hooke's law, this paper establishes theoretical models for a tight-jacketed sensing cable and a reinforced sensing cable based on elasticity mechanics and deduces the relationship between the strain of the cable and its stress. Then it does the simulation and experiment on samples of these two typical types of sensing cables by using Brillouir Optical Time-domain Reflectometer(BOTDR) to detect the strain of the cables. The results of theoretical calculation and the experimental measurement show that the linear rates of all the results are more than 97%, and the differences of the slope between the theoretical calculation and the experimental measurement are less than 8%, which proves the models to be effective.The models can provide a basis for stress sensing systems in determining the stress situation for an objective.

The detecting performance of an airborne daytime star sensor for navigation is analyzed and a daytime star sensor prototype is researched. Mathematical models of Signal to Noise-Ratio(SNR) and contrast of a photoelectric imaging system in daytime under noises are established, and calculated results by different focal lengths and different apertures are compared to conclude the influence of system parameters on the detection results and the requirements of optical system.The conclusion shows that the proper relative aperture of optical system is 1/15. A daytime star sensor prototype for tracking stars of magnitude 3 is designed. Tracking images of the polestar from sunrise to noon are acquired by a ground experiment using a F/15 objective, and brightnesses and contrasts of star images at different time are compared. It shows that the contrast of star image has descended from 1.55 at 5:00 a.m. to 1.2 at 12:00 a.m. The minimal degree of modulation is more than 0.03 by calculating with the prototype′s parameters, which shows that the optical system with Φ70 mm and F/15 can satisfy the detecting requirements. The theoretic analysis and design is proved to be available, and the research can be a reference for developing of miniaturized and precise airborne daytime navigation star sensors.

Milling tests for SiCp/2009Al composites were performed by using Polycrystal Diamond(PCD) tools at the cutting speed of 600-1 200 m/min and the mechanisms of tool wear were investigated. A Scanning Electron Microscope (SEM) was used to examine the machined surfaces and tool wear land, an X-ray diffractometer (XRD) was used to analyze the substances on the machined surface and the laser Raman spectra and Energy Density Spectrometry(EDS) were taken to analyze the elements on the tool wear land. The results show that the tool grain breaking-off, abrasive wear, chipping and the peeling caused by the high frequency impact and the scrapt of SiC particles are the prevalent wear patterns. When high volume fraction materials or heat treated materials are milled, or the milling is at a higher cutting speed, micro-cracks will form on the PCD tools. Furthermore, volume fractions, average sizes of SiC particles, tool grain sizes, cooling and heat treating conditions have significant influence on the tool wear. It concludes that the life of PCD tools is in the range of 70-240 min. The micro-cracks are produced on the tool flank under the combined effects of SiC particle impact, cutting vibration and heat impact. Moreover, the aluminum and copper can be diffused into the tool matrix under the very high cutting temperature and cutting forces and the trivial graphitization is taken place on the PCD tools due to the catalysis of copper in the aluminum matrix.

The relationship between the dynamic effect of a capillary and the length of microchannel in front of a step valve was researched, then the influence of dynamic effect in the process of stopping the capillary in microchannel formed by hydrophobic and hydrophilic walls on the step valve was analyzed. According to the working principle of the passive valve of capillary and energy conservation law, the critical length in front of the step valve was calculated when the step valve stoped the capillary. The calculated length corresponding to the critical length was obtained by numerical simulation. It shows that when the actual length in front of the step valve is greater than or equal to the calculated length, the valve can stop the capillary effectively. The step valves were fabricated by bonding polydimethylsiloxane (PDMS) to glass. Experiments of stopping the capillary by the step valve were performed in a rectangular microchannel formed by three hydrophobic PDMS walls and one hydrophilic glass wall. For the series of microchannel with a depth of 40 μm and widths from 200 μm to 400 μm, the calculated length in front of the step valve is from 4.531 6 μm to 10.081 μm. In the experiments, the actual length before the step valves for effectively stopping the capillary is in the interval from 10 μm to 2 000 μm. The results show that even if the actual length in front of the step valve is 10 μm, the step valve can also stop the capillary. Therefore, the dynamic effect of the capillary can be ignored in stopping the step valves in the microchannel.

In order to increase accuracy and reduce temperature drifts of pressure sensors, a micro-machined resonant pressure sensor with H-type doubly-clamped lateral beams was proposed based on electromagnetically driving and differential detection. First, Finite Element Method(FEM) analysis was performed to guarantee the sensitivity and the resolution of the sensor. Then, based on the boron-diffusion self-stopped etch technique, a sensor sample with good uniformity was achieved by a standard bulk Micro-electro-mechanical System(MEMS) process. Finally, the vacuum package of the sensor was accomplished by adhesive bonding using non-photosensitive BCB(Benzocy-clobutene, DVS-BCB-3022-46) and stress isolation. Experimental results show that the nonlinearity is lower than 0.02% in 0 to 120 kPa with a accuracy of 0.05% FS, and the temperature drift is less than 0.05%/℃ in -40 ℃ to 70 ℃.The sensor can suppress temperature drifts and realize pressure measurement in wide measuring range and higher precision.

As the SU-8 cantilever integrated by a metal piezoresistance can achieve a very high force sensitive coefficient due to its low elastic modulus, this paper designs a new type of SU-8 cantilever micro-force sensor integrated by a serpent-shaped copper piezoresistive structure and fabricates a prototype with a double-layer cantilever by a novel processing method. It introduces the design principles and fabrication method for the micro-force sensors and measures its technological parameters.Experimental results show that the SU-8 micro-force sensor has good linearity range in 0-350 μN and its force sensitivity is 0.24 mV/μN and measuring error is 4.06%. It can implement the micro-force detection and has advantages of simpler production process and shorter cycle as compared with a silicon micro-force sensor. Moreover, the force sensor has potential application in biomedical research for the SU-8 with prominent biological compatibility.

As material characteristics of a bond layer directly affect the measurement accuracy of Fiber Bragg Grating(FBG) strain sensors, this paper simulates the impact of elastic modulus of the bond layer materials on strain transfer coefficient K according to the bond layer strain transfer model and the formula of strain transfer coefficient K under the premise of determining the structure size parameters. The results show that the strain transfer efficiency of metal is superior obviously to that of the organic adhesive.A contrast test is carried out. Test results indicate that the strain transfer coefficient K of the lead maintains a level of 0.98 while that of epoxy maintains about 0.90, and the former′s transfer efficiency has increased by 8.9% as compared with that of the latter.By replacing epoxy with the lead, it can greatly improve the strain measurement precision, and can more fully exert fiber optic sensors′ strain measurement ability.

In order to improve the measuring accuracy of a steel ring reflective grating, a grating signal compensation system is designed. The system is used to compensate the signal amplitude, DC voltage and the phase of the grating, respectively, for which two grating signals have a better orthogonality. First, according to the sampling method for steel ring reflective grating signals, an electronic processing system is created. And then several compensation algorithms are established aiming at the signal amplitude, DC voltage and the phase.By taking the steel ring reflective grating with 8 192 lp/mm as a platform and the results of testing precision as a basis of adjudication, the proposed signal compensation system is verified. The results show that the precision compensated has been improved by 2.28″ with an increasing magnitude of 50%.Obtained results prove that the signal compensation system is feasible.

A pre-stressed six-axis force/torque measurement method based on a Stewart platform was established to extend its measuring range. Compared to the traditional Stewart six-axis force/torque sensor, the principle of the pre-stressed Stewart six-axis force/torque measurement was analyzed. Then, structure design of the sensor was finished. By using the ANSYS software, a finite element model for the pre-stressed Stewart six-axis force/torque sensor was established. According to this model, the vibration modal was analyzed, and the natural frequencies and vibration modes were obtained. Finally, principle of the sensor calibration was analyzed, and the sensor coupling matrix was calculated according to the data of a static calibration experiment. Experimental results indicate that this pre-stressed Stewart six-axis force/torque sensor has the measuring ranges of 0～3 000 N for the force and 0～300 N·m for the torque, and its measuring accuracy is better than 7.5% of the actual data. It shows its advantages in wide measuring range, easy to decoupling and convenient for assembly.

To ensure high-precision calibration of gear measuring instruments, the international standard ISO / TR 10064-5:2005 recommends a Double-Ball Artifact (DBA) as calibration reference. The DBA employs a high-precision sphere to replace the involute and uses an error compensation technique to obtain a high-precision involute to realize the involate traceability. The theoretical issues relating to the DBA are discussed to provide a theoretical basis for the design and application of the DBA. The error of measurement principle of DBA and its calculation function are given under the conditions of ideal installation and with installed errors for the electronic generative metrology. According to the optimized arc thoughts, the formula of the optimized center distance between the two balls is derived by the least square method. Finally, the error compensation method is described.The DBA software is designed based on the above theory and a DBA is developed with a parameter of module in 2.268 1 mm, number of teeth in 40, and a pressure angle in 20 °.

A design method for the digital camera based on a back illuminated 1 024 pixel×1 024 pixel Electron Multiplying(EM)CCD image sensor is proposed. The Complex Programming Logical Device(CPLD) is adopted to generate the CCD driver timing and video processing timing, EL7156C pin driver chip is used for standard EMCCD clocks, and a discrete circuit is used for EM clock. By adopting a CCD analog front-end chip VSP2566 with Correlation Double Sampling(CDS) function to sample, the image data of the CCD output are digitized. The design methods of logic timing control, driver circuit, and video processing circuit are introduced. Experiments show that the digital camera circuit can output 16 bit digital image data simultaneously at a speed of 8 frame/s, its optical dynamic range can exceed 70 dB, and the remote programmable control electron multiplication can operate at the gain more than 50 dB. The design has been used in the remote EMCCD imaging system, and also can be applied to the low light level imaging area and radiation diagnosis field.

This paper attempts to improve the commonly used image multi-focus fusion methods, for they could not identify meaningful image features from noises. An antinoise multi-focus image fusion algorithm is presented. The improved adaptive block-based image fusion algorithm combined with a new focus measure with noise immunity is used to focus the noisy image effectively and to achieve good fusion results. Root Mean Square Error (RMSE) and Mutual Information (MI) are selected to evaluate the fused noisy image with different intensities and comparison experiments are performed.As compared with those of contrast pyramid, wavelet transform and Contourlet transform, the average RMSE of the fused image by the proposed method has been decreased by 4.288 9, 4.479 1 and 4.187 1 respectively, while the average MI increased by 2.366 4, 3.282 5 and 2.063 9, respectively. With the noise interference, the proposed method can maintain the useful information of the source images accurately, suppress noise effects effectively and obtain better image fusion quality.

Infrared Focal Plane Array (IRFPA) is easy to produce a fringe nonuniformity (NU) when it works for long hours or in a changed environment. Aiming at this problem, a dynamic Nonuniformity Correction(NUC) method was proposed in no covering imaging fields and its hardware realization technology was investigated. A scene-based algorithm, Improved Constant Statistics Nonuniformity Correction (ICS-NUC) was presented, in which the fringe NU was reduced by balancing the statistics of vertical channels, and a dynamic NUC was realized by using a forgetting factor in inter-frame iterative process. The ICS-NUC algorithm flow based on an uncooled IRFPA module with BF561 DSP platform was described. Then, two ICS-NUC algorithms, ICS by Average (ICSA) and ICS by Median (ICSM), were implemented on the platform. By these ways, the fringe NU was corrected dynamically, and whole the infrared image quality was improved. The experimental results show that the fringe NU of IR image is reduced after ICS-NUC and the NU of IRFPA module is decreased from 3.43% to 1.82% by ICSA and 0.91% by ICSM, respectively. After continuous operation by four hours, the NU of IRFPA module is still decreased from 5.05% (drifted) to 0.92% by using ICSM.It concludes that the realization of ICS-NUC algorithm on IRFPA module can lay a technical foundation for the following study and applications of high-performance thermal imaging systems.

A spot centroiding algorithm immune from noise pixels is proposed to improve the existing centroiding algorithms that can not process images with discrete noise pixels directly, and it is achieved by a Field Programmable Gate Array(FPGA). This algorithm does not depend on noise pre-process or post-process to eliminate noise pixels. Firstly, the algorithm marks the background pixel, noise pixel and the spot pixel, respectively, then each pixel is marked after it is compared with the neighbouring one. At the same time, centroid parameters that belong to the same spot are accumulated, while the real noise pixel is bypassed. Compared with currently existing algorithms, this method takes full advantage of the parallel processing ability of the FPGA. It can extract spot centroid coordinates and eliminate the noise pixel simultaneously when the image pixel is output, the pre-process image is not needed and the storage space is saved. This algorithm is especially suitable for processing the spot image with discrete and high-brightness noises caused by long exposure time.

In order to balance the robustness and building complexity of a feature descriptor, a local feature description algorithm based on Laplacian is presented. It is analyzed and illustrated that the Laplacian not only has good properties to Euclidian transformation, zoom, and linear brightness changes of an image, but also can characterize the local structure of the image. On the basis of that, a 64-dimension descriptor is built with the response of Laplacian of Gaussian. Finally, the descriptor is used to match feature points with the absolute distance as similarity measurement. Simulation results indicate that the proposed descriptor can obtain better matching results for the image with zoom, rotation, blurring, illumination varying as well as smaller viewpoint changes, and the matching speed is more than 4 times that of Scale Invariable Feature Transformation(SIFT). The proposed feature description algorithm is suitable for matching the images of structured scenes, for it is insensitive to the image transformation with rotation, zoom, luminance varying, compression or small viewpoint changes.

An integrated Micro Attitude Heading Reference System(MAHRS) based on Micro-electro-mechanical System(MEMS) gyroscopes, accelerometers, and magnetic-sensors was researched for Micro Aerial Vehicles (MAV). As traditional algorithms could not keep the attitude accuracy of the MAV during a long-time maneuvering, an adaptive attitude estimation filtering algorithm with low-cost multi-sensors was presented for the MAHRS. Firstly, this algorithm was used to establish the models of acceleration and magnetic-disturbance and took them into a state equation. In this way, it could maintain the attitude accuracy of loads during the long-time maneuvering. Then, this algorithm adopted the federated filter mode to reduce the interaction between acceleration and magnetic disturbance to improve its precision and reliability. Moreover, the algorithm made the estimation variance (P) and measurement variance (R) remain steady adaptively during different maneuvering states. By comparing with other different algorithms for the MAHRS, It shows that the presented algorithm is more accurate and reliable than other algorithms during various maneuvering states.

In order to improve the dynamic measurement range and measurement accuracy of the Hartmann-Shack Wavefront Sensor (HSWS) in an adaptive optics system, the centroid detection for a wavefront spot pattern collected by a CCD is researched. The principle of HSWS is introduced and an adaptive centroid detection method by dynamic positioning spot region, dynamic segmenting threshold level and locking optimal detection window is presented according to the structure features of HSWS with 127 units and the distortion characteristics of actual human eye spot pattern.The effects of threshold level selection on human eye spot centriod detection are discussed and the centroid detection accuracy and noise robustness of the method are analyzed through simulation and experiments. Experimental results indicate that selecting the threshold level to be (50±3)% of the total grade amount is more appropriate and the centroid detection error can be decreased by 60% or more as compared with that of other detection methods. By using adaptive centroid detection method in the HSWS of an adaptive optical retina imaging system, human eye aberration can be reduced from 0.728λ to 0.081λ in RMS(λ=785 nm) through a closed loop correction.Moreover, the system can reach the diffraction limit basically and can obtain a retina image. These results show that the adaptive centroid detection method can select different experimental parameters aimed at the characteristics of spots, which makes up the limitations of the general methods.It meets the feasibility and practicability of human eye aberration measurement.

To improve the remote sensing image classification accuracy by incorporating labeled and unlabeled samples, this paper proposes a new manifold learning method called Semi-supervised Manifold Discriminant Embedding (SSMDE). This method uses data point labels to construct two relational graphs, within-class graph and between-class graph, they then are taken to encode the class relation information indicated in the labeled data points and to construct two weighted matrices. The labeled and unlabeled data points are utilized to construct the total scatter matrix to describe all the data points. Finally, the projection matrix of SSMDE is obtained by solving an optimization problem. The SSMDE method can not only take into account the discriminant information of labeled data, but also preserve the global structure of all data points. The experimental results on both synthetic and remote sensing images show that the proposed method can achieve the classification accuracy of 92.32% and the error between the classification results by the SSMDE and the government statistics is less than 5%, which demonstrates the effectiveness of SSMDE.

A real-time imaing nonuniformity correction algorithm on a Field Programming Gate Array(FPGA) platform was proposed to modify the imaging nonuniformity of a multi Time Delay Intration(TDI)CCD mosaic camera. Firstly, the definition of Photo Response Nonuniformity(PRNU) and its generation in the multi-TDICCD mosaic camera were introduced. Then, the synthetical correction algorithm was presented, in which the two-point correction method was used in a single channel, gain averaging correction method among multi-channels and the scene-adaptive correction method among multi-TDICCDs. By considering the reliability and the influence of finite word-length of the FPGA, the synthetical correction algorithm was optimized and implemented on the signal processing FPGA platform. Furthermore, a scheme for modifying correction factors and spaceborne calibration were given. The experimental results indicate that the imaging nonuniformity can be decreased from 4.82% to 0.27% for a single TDICCD, and to 0.41% for a 3-TDICCD mosaic camera with the proposed correction algorithm.It also proves that this correction algorithm has high real time performance, good practicality and efficiency, and satisfies the requirements of target projects.

Sub-pixel positioning technology for CCD images is proposed to improve the astronomical location precision of space targets.The main factors affecting the positioning accuracy of targe centroids, such as image noise, discrete sampling error and window size selection are analyzed. Combining with the CCD imaging characteristics for dispersion spots, a location method for target centroids based on energy accumulation is proposed. This method uses the interpolation technique to reduce the illumination inconsistency between actual sensitive area and discrete sample points, and uses the energy accumulation of dispersion spot to select the efficient target window for optimizing target gray and eliminating random noises. Finally, a squared weighted centroid localization method is taken to calculate the centroid position. Experimental results show that the method has strong anti-interference ability and stability.It can offer a positioning accuracy of 0.01 pixel and is suitable for a low SNR condition.

An iterative blind image restoration algorithm based on Self-deconvolution and Incremental Wiener Filter (SDIWF-IBD) is proposed. The self-deconvolution estimation for a Point Spread Function(PSF) is applied to Iterative Blind Deconvolution(IBD) to estimating exactly the frequency domain of the PSF. The incremental Wiener filter is used in the image estimation of IBD to keep the algorithm convergence steady. To further control the convergency, an in-iterative acceleration is suggested to control the speed of algorithm convergence and reduce total external iteration. Experimental results indicate that more details are recovered in the restoration image with few distortions, and the algorithm is converged to a small error quickly. It concludes that the SDIWF-IBD algorithm has good restoration ability at a fast and controllable convergency speed, and is fit for applications in real-time.

A target recognition method based on local fuzzy thresholds is presented to solve the problems of serious background interference, the absence of reference map for complex terrain objects and the low contrast between the target and the background. Firstly, a multi-threshold algorithm is designed and a saliency map is produced based on the establishment of multi-scale space. Then, the Itti model is improved and a candidate target filtering model is constructed by using the local fuzzy threshold method based on an image fuzzy rate. Finally, the detected results are precisely matched using Integral Nprod to determine the right one. The experimental results indicate that the match rate of the algorithm has increased nearly by 20% and 40% and the time consumption by 75% and 50% as compared with those of Hausdorff distance algorithm and Integral Nprod algorithm, respectively.In conclusions, the new algorithm has the advantages of high match rate, high speed and high accuracy for FLIR targets in complicated backgrounds.

Traditional Simultaneous Localization and Mapping(SLAM) algorithm is lack of the ability to describe multiple sensor information accurately in a clutter environment, and it is prone to false data association. Therefore, this paper proposes a SLAM algorithm based on Probability Hypothesis Density (PHD) filter to deal with these problems. By taking the sensor observation and environmental map as random finite sets in every time step, a joint target state variable is constructed. Then, with the Probability Hypothesis Density(PHD) filtering, the poses and environmental map of the robot are estimated simultaneously and the PHD filter is realized by a particle filter. To avoid the error caused by cluster, a time-delay particle set outputting approach is proposed for joint target state extracting. The new algorithm can describe the observation uncertainty, loss detecting, false alarm due to a clutter and other sensor information accurately, and also can avoid the data association, by which the system state estimation is closer to real values. The simulation results show that the accuracy of the new algorithm in the vehicle localization and mapping is improved by more than 50% as compared with that of traditional SLAM algorithm. It provides a new solution for SLAM problems in the clutter environment.