On the structure of optical system, the causes of stray light of a rigid endoscope and its suppression methods were explored to eliminate the stray light and to improve its imaging performance. Firstly, the stray light in the optical and mechanical system was simulated and analyzed by ray tracing in Lighttools. By analysis of the stray light path, it points out that the stray light of the optical system comes from the total reflection of a relay system and an objective system. Then, with the possible causes, the suppression methods for the stray light were proposed, including adjusting objective structure and controlling ray path. Finally, a rigid arthroscopy with a Field of View(FOV) of 70° and an outer diameter of 2.7 mm was designed to suppress the stray light. With proposed methods, the Modulation Transfer Function(MTF) reaches diffraction limit, the values of all fields are higher than 0.4 at 100 lp/mm, and the size of spot diagram is smaller than 4 μm. Besides the high-definition imaging, simulation demonstrates that the stray light has been moved out entirely in the optical system, and the system shows excellent imaging quality.

To obtain a 3D model of microscopic dynamic samples, a 3D microscopic imaging method based on a single objective and red-blue dual-apertures was proposed and a corresponding 3D microscopic specimen system for capturing images was established. In the system, the dual apertures covered with a red and a blue filters were respectively placed in the exit pupil plane of the objective. When it was illuminated by a white light, the color CCD in the imaging plane of objective could receive the perspective views of specimen expressed by red and blue images respectively at one exposure. The stereo image with an optical parallax for the specimen could be obtained by digital image processing, and the 3D model of the specimen could be further reconstructed by a phase matching algorithm. Experimental results indicate that the system has reconstructed the 3D model with a Central Depth Plane (CDP) up to 241.50 μm.As compared with double-objective 3D microscopic imaging systems, the proposed system is more compact and simpler and can obtain a 3D model of microscopic dynamic sample at one exposure, so that the system has important application prospects on microscopes and endoscopes.

According to the demand of a high precision optical fiber gyro for the light source′s stability, the digital thermal control scheme for an erbium-doped super fiber source(SFS) was proposed based on the 32-bit digital signal processor TMS320F2812. By taking the erbium-doped SFS as the study object, the advantages and disadvantages of the existing light source temperature control technologies were analyzed, and the program of the "numeric constant current source + digital thermostat" method was developed according to the analog control scheme. The operating characteristics of a Thermo Electric Cooler (TEC), the internal structure and heat transfer mechanism of the SFS pump source were studied, and a mathematical model of the die temperature control system of the SFS light was established. Furthermore, a corresponding continuous domain lead-lag correction network was designed, the controller was discretized and a PID digital compensation control algorithm was realized. Finally, the control precision of the digital temperature control system for the SFS light source was verified experimentally. Obtained results show the precision is better than ± 0.05 ℃ in 20-90 ℃ and meets the requirements of optical fiber gyros for low power consumption and miniaturization.

Elastic backscattering spectroscopy is an effective approach to study the particulate composition of biological cells. This paper explores the experimental methods to measure and obtain the elastic backscattering spectra and their advantages and disadvantages. Regarding the microsphere measurement, the pros and cons of two popular spectral acquiring methods are experimentally compared and analyzed, including the method based on a fiber probe and that based on a lens system. The experiments demonstrate that the cross-correlation coefficient between measured spectra by lens system and the Mie calculation is 0.96, higher than 0.93 that is by the fiber probe under an optimum condition, because the lens system results in a well-defined scattering angle for the acquired spectrum. Therefore, it points out when a higher precision in particle sizing is required, the lens system is a better choice and also the filtering denoising will be used to improve the precision of particle sizing. As the spectrum capturing by the optical probe is simple and easy to be implemented, it can be used in a spectral capturing needed by a lower precision.

A high precision inspection method based on image mosaic was proposed for the final optics inspection in an Inertial Confinement Fusion (ICF). The image mosaic method contains image registration stage and image fusion stage. In the image registration stage, a registration method based on feature matching of local damaged area was used to overcome the exposure difference between the subimages and the lack of image information, and the moving information in the inspection system was used in image matching to improve the inspection precision and inspection efficiency. In the image fusion stage, the Poisson image fusion was adopted to eliminate the "stitching line" caused by an exposure difference and to implement the seamless mosaic. The images of final optics inspected by Final Optics Online Inspection (FOOI) was compared with the proposed method and Scale Invariant Feature Transform(SIFT) method,and the results show that the proposed method is superior to SIFT method. The 4×4 mosaic image shows that the inspection resolution of the FOOI device in the ICF has been improved significantly by proposed method.

To improve the measurement accuracy of a plane grating diffraction efficiency measuring instrument, the affect factors on the diffraction efficiency measurement of a plane gating were researched. According to the measuring principle of relative diffraction efficiency of the plane grating, the concept of beam cross-section factor k(θ) of plane grating was proposed and the correspondent expression was developed. Then, the factors affecting the measurement accuracy of diffraction efficiency, the spectral width and the beam cross-section factor were analyzed, and two independent variables for the regression were defined by combination with the measurement value of diffraction efficiency. Regarding the nonlinear relation between the theoretical value of dependent variable and the independent variable defined, the complete quadratic regression analysis was applied as the methodology to establish a revised equation to improve the measurement accuracy of the diffraction efficiency. The results show that the variation between the revised and the theoretical values is less than 0.5%, which means the measurement accuracy of diffraction efficiency has been improved and the performance of the instrument is enhanced significantly.

As the precision of roll angle of a star sensor is much lower than that of pitch/yaw angles, it has great impact on the course precision of a space tracking ship. By considering the actual working environment of the space tracking ship, an efficient ship attitude determination method based on optical axis pointing of a dual star sensor was proposed. Firstly, the principle and the shortages of ship attitude determination method based on the single star sensor were presented, and the impact of roll angle of star sensor on ship attitude measurement precision was analyzed. Then, a new attitude measurement system based on dual star sensors was constructed, and its calculation model based on optical axis pointing was built. Finally, the outdoor experiments were carried out to verify the precision of the algorithm model. Experimental results indicate that the accuracy of ship attitude based on dual star sensors can reach to 6.9″, 5.7″and 4.5″ for the course , pitch angle and the roll angle, respectively. Comparing to the single star sensor, the precision of ship attitude determination is improved greatly, and the impact of roll of star sensor on course of the ship is avoided, which offers useful references for engineering application of star sensors used on space Tracking Telescommanding and Control (TT&C) ships.

A new method to test the optical performance of a medical rigid endoscope was proposed to inspect its quality. First, the mesh maps of images were transformed and qualitatively observed and judged by a quality testing device. Then, the images were treated using gray-scale transformation and binarization to find the chord-line passing through the center and to draw up plot curves. By fitting analysis of plot curves, the horizontal and vertical ratio of edge and center and the FOV difference in brightness ratio were both calculated. Obtained data were analyzed and experiments on the domestic and overseas medical rigid endoscopes were performed. The results indicate that the FOV difference in brightness ratio of domestic endoscopes has a good correlation in horizontal and vertical directions, their values are all above 0.908 5 and the correlation is independent on the brightness. The imported endoscopes have a better relevance ratio of edge and center in horizontal and vertical directions and their values are above 0.656 5. Moreover, FOV difference in brightness ratio is proportional to brightness, and the ratio of edge and center is stable. It concludes that the chord-line method is able to calculate the FOV difference in brightness ratio and also shows the performance evaluation vividly. This new method is simple and convenient, and suitable for test and evaluation of the light efficiency performance of medical rigid endoscopes.

A method by using magneto-optical modulation and light source modulation techniques was proposed to measure the extinction ratio of a polarizing prism precisely, and an extinction ratio measurement system was also established based on the two modulation techniques. Firstly, the system′s measurement model was derived according to the Jones matrix describing manner of polarized light. By using a magneto-optical modulator, the optical axis angle between a polarizer and a measured polarizing prism was precisely located. Then, by using a chopper to modulate the square wave of a light source, the impact of noises on the system measurement accuracy was eliminated and the system could measure the light intensity value accurately after the measured polarizing prism′s axis was located precisely. Finally, the polarizing prism was measured for a number of times and the data were averaged. The experimental results indicate that the measurement accuracy of extinction ratio for the polarizing prism can be 10-6, which verifies the effectiveness and stability of the method. The system is characterized by high stability, high accuracy, and can offer the references for the performance test and engineering applications of polarized devices.

A new type of optimized Photonic Crystal Fiber(PCF) with high birefringence and high nonlinearity was proposed by adjusting its structure parameters. The cladding of the PCF was arrayed by circular air holes of different sizes in accordance with the hexagonal lattice and two ellipse air holes were guided to its fiber core centre along the x axis. By using the full vector finite element method with anisotropic perfectly matched layers, the mode field distribution, effective index, effective mode area, nonlinearity and birefringence and the fiber structure were investigated simultaneously and the effects of different optical parameters on the optical characteristics were analyzed. The analysis shows that the birefringence and nonlinearity properties can be controlled by adjusting the sizes of large air holes and ellipse air holes. The numerical results indicate that a better result can be obtained when the parameters of the PCF are adjusted appropriately.The maximum nonlinear coefficient of 53.5 W-1·km-1 and high birefringence of 1.092 9×10-2 are obtained at 1.55 μm, in which the birefringence is two orders of magnitude higher than that of the traditional optical fiber. The modified fiber has a wide application prospect in the communication system that needs high nonlinear and high birefringence.

A novel magnetic field sensor based on a chirped effect of Fiber Bragg Grating(FBG) was proposed. The relationship between the magnetic induction intensity and the variable bandwidth of reflection spectrum of the FBG was derived. The working principle of the sensor was introduced.Firstly, the magnetic force from an energized spiral coil was applied to a disc-shaped soft iron in a magnetic field to make the rectangular cantilever beam deformed, which resulted in the bandwidth change of the reflection spectrum of FBG that was mounted at a side of the cantilever. As mentioned above, the magnetic induction intensity could be obtained by measuring the variable bandwidth of the reflection spectrum of the FBG with a spectral analyzer. The experiments show when the optical spectral analyzer is with a resolution of 0.001 nm, the measuring range of the sensor is up to 6-70 mT. Experimental results demonstrate that the variable bandwidth of the reflection spectrum of FBG is not sensitive to temperature change. The 3 dB bandwidth change is less than 8 nm when the temperature changes from 0 to 45 ℃, which is consistent with theoretical value well. The sensor scheme is feasible and practical.

A precise focusing technology for a spliced TDICCD camera was introduced. Firstly, the components of focal plane were assembled to the designed locations on the camera under the circumstance of a constant temperature and vibration-isolation. Then, the Modulation Transfer Function(MTF) of the camera image was obtained through a data acquisition and processing system, the curve of MTF-focus value of every TDICCD image was plotted, and the optimum imaging plane of every TDICCD was obtained. Finally, the common optimum imaging plane of the 8 chips for TDICCD was fitted according to the every single optimum imaging plane obtained above, and then the machining values of the external interfaces of the focal plane components were derived to finish the assembly and adjustment. Precision analysis on the focusing shows that the focusing error is 9.1 μm and the relative error to the depth of field is 4.4%. The MTF test indicates that the 8 chips of TDICCD have approximately even MTF values, which are high than the camera′s static MTF requirement of 0.2. Obtained results coincide with the variation regularity of the camera′s optical MTF values and can provide references for the precise focusing technologies of the same kinds of remote sensing cameras.

To fabricate deep and small holes on the difficult-to-cut metals, an electrochemical drilling technology with chip removal and electrolyte replenishment was proposed. It was named electrochemical drilling of deep and small holes with high speed micro electrode. The working mechanism of the technology was analyzed and researched and the influences of some predominant parameters such as rotary speeds, voltages, pulse frequencies, and machining speeds on machining precision and stability were investigated. It points out that high machining precision and stability can be obtained by matching above parameters reasonably under a higher machining efficiency . A groups of deep holes with the diameter less than 0.5 mm and the aspect ratio more than 10 were fabricated successfully on the nickel base super-alloys GH4169 by using the hard alloy micro spiral electrode in a self-developed high-precision micro-electrochemical machining system. These holes show good morphology, lower tapers and clear-cut margins. It is proved that the high-speed electrochemical drilling process for fabricating deep and small holes has a huge potential and broad application prospects.

A modeling method and a real-time trajectory tracking control strategy were proposed based on Hammerstein model to overcome the effect of rate-dependent hysteresis on piezoelectric actuators. First, by using modified Prandtl-Ishlinskii(MPI) model and the Autoregressive Model with Exogenous Input(ARX) to represent the static nonlinear part and the linear dynamic part of the Hammerstein model, respectively, a Hammerstein model was proposed to describe the rate-dependent hysteresis of piezoelectric actuators. Then, based on the proposed Hammerstein model,a compound control strategy with feed-forward adaptive inverse compensation and PI feed-back was designed. Finally, to verify the effectiveness of the proposed control strategy, a compound control strategy with feed-forward inverse compensation and PI feed-back was designed to compare with the proposed scheme. The experiments on the real-time tracking for single frequency signals and sweep signals with the amplitudes of 11 μm, compound frequency signals and sinusoidal scanning signals with variable amplitudes were performed in a frequency of 100 Hz, obtained results show that the root mean square error and the relative error between the reference input and the output of piezoelectric actuator are 0.280 8-0.437 3 μm and 0.016 5-0.024 4 respectively. As compared with the compound control strategy with feed-forward inverse compensation and PI feed-back, the proposed control strategy has better tracking accuracy and real-time ability.

The integral Windup phenomenon from a position angle control system in the aviation camera is more likely to happen when a big travel is given, which will affect the dynamic performance of the camera. To improve the dynamic response performance of the position-angle control system on this condition, this paper proposes a method based on the Anti-Windup variable structure adaptive PID control to design the position angle control system. A primary and secondary cascade dual loop was used to control the position of a position-angle mirror, and a variable structure adaptive PID control based on the Anti-Windup was adopted to control the compensation in the speed inner loop. The PID tuning method based on real-coded genetic algorithm was taken to tune the initial PID control parameters of speed inner loop, and the optimal index of parameter selection used punishment function to eliminate the overshoot. As mentioned above, the position- angle control system is able to achieve satisfactory performance in a large position given. Experiments show that the control strategy based on the Anti-Windup variable structure adaptive PID control avoids the position angle mirror coming into a saturation in the back stage after auto-focusing, increases the adjust time of back stage by 35.7% and improves the return performance of the position angle mirror.

To reduce the driving voltage of the digital microfluidic device, the traditional square electrode structure was designed to crescent shapes, and the driving effects of crescent shape electrode structures with different parameters were investigated. First, the effect of different electrode shapes on reducing the driving voltage was analyzed based upon the theory of electrowetting-on-dielectric. Then, the moving process was numerically simulated by means of Volume of Fluid (VOF). Based upon simulation results, the moving process of the same droplet in different electrode structures was analyzed. Finally, four electrodes with different crescent shapes was designed and their driving effects for the same droplet were verified. The experimental results show that the crescent electrode device with the arc diameter equal to the length of the electrode is able to reduce the driving voltage by 15.6% more than those of other three crescent shape electrode devices. Besides, the 1 μＬ droplet on this electrode structure device can be driven successfully with the velocity of 1.6 cm/s when the driving voltage is just 16 V, accordingly, which shows that this configuration design is the optimal structure among various crescent electrode structures. In conclusion, the obtained experimental results validate the feasibility that the crescent electrode with the arc diameter equal to electrode length can reduce the driving voltage effectively.

To perfect the simulation model of a piezoelectric unimorph cantilever and to design optimal piezoelectric vibrators, the analysis model of equivalent viscous damping coefficient for the vibrator was established according to the mechanical vibration theory and damping theory. Then, the influences of the material properties, structural dimensions as well as the cross-section shapes on the machinery and electricity damping coefficients were analyzed. Finally, three groups piezoelectric vibrator samples with different shapes and sizes were prepared , and the impact vibration test were performed to verify the theoretical analysis. The results show that the impact of piezoelectric material layer on the overall damping depends on the elastic modulus ratio of the substrate and the piezoelectric material. In each vibration cycle, the piezoelectric material power loss is proportional to the cube of the cantilever length and inversely proportional to the width. However, the damping ratio to determine an amplitude amplification factor is not varies with the structure size monotonically. The difference beween the experimental results and theoretical model is 2.5%-14.7%,which proves the reliability of the theoretical model. Moreover, the piezoelectric unimorph cantilever with optimal static characteristics may not have the best dynamic characteristics under the same extreme load. These results can offer reference for the optimal design of piezoelectric vibrators.

By taking a long focal length oblique CCD camera as an object of study, a multiple loop control scheme based on the high gain observer was proposed to improve the position accuracy of auto-collimating auto-focusing lens barrel in a high dynamic state. Firstly, a mathematical model was established according to the open-loop amplitude-frequency characteristic curve of a depression angle control system. The feasibility of control scheme was analyzed by the control theory and the simulation tests were adequately performed. Then the reliability of control scheme was further validated through the practice control system. The experimental results show that the high gain observer can estimate the angle speeds and angle acceleration well. As compared with the double closed loop control method, this proposed control scheme reduces the maximum position control disturbance error by 96.3% and ensures the requirement of lens barrel positioning precision within ±0.05° .Hence the new control method is valuable in practical engineering applications.

On the basis of higher-order resonant characteristics of silicon cantilevers of Atomic Force Microscopes (AFMs), a high speed scanning method for dynamic AFMs based on the higher-order resonant cantilever was put forward, and an AFM working at one-order resonant and higher-order modes was developed. The basic structure and working principle of the higher-order resonant AFM system were introduced and the feasibility of the method by using the higher-order resonant characteristics of cantilever to realize high speed scanning was demonstrated theoretically. With home-built AFM as the investigated object, the main factors influencing the scanning speed of the dynamic AFM were investigated, and the response time of each system module was analyzed and estimated by tests. It is experimentally proven that the settling time of the second-order resonant mode AFM is less than that the first-order resonant mode AFM obviously. Finally, the same area of a grating sample was scanned by the first-order and the second-order mode AFMs respectively and the experimental results demonstrate that the scanning speed of the second-order mode AFM is about 3.3 times faster than that of the first-order resonant mode under the same condition. Theoretical analysis and experimental results prove the feasibility and effectiveness to improve the dynamic AFM scanning speed by using the higher-order resonant cantilever.

As the common altitude sub-assembled optical window for a aerial camera with larger view angle has a lower gradient, this paper proposes a protector for the altitude sub-assembled optical window. The mission requirement of the window and its working environments were analyzed, and the protector was designed. A transmission mechanism in the protector was designed and the transmission accuracy was analyzed. Finally, the vibration test and simulated altitude environment test were performed. The analysis and test results indicate that the first natural frequency of the protector is 111.9 Hz, which can ensure the dynamic responce of the protector under dynamic loading. Moreover, the movement of transmission mechanism is stable and reliable in low-temperature (-56.5℃) and low-pressure (5.5 kPa) environments, and the speed error is 1.3%. The protector has been successfully applied in an aerial camera.Experiments show it has effectively overcome the impact of altitude aerodynamic resistance and low-temperature, low-pressure environment, and can be used as a reference for protector designs of other sub-assembled optical windows.

To track and control the trajectory of a Wheeled Mobile Robot (WMR) in the smooth, robust and stable modes, the principle of the bio-inspired dynamics was analyzed, the nonlinear Model Predictive Control (MPC) was explored and a MPC approach based on bio-inspired dynamics was proposed. Firstly, a bio-inspired dynamics sub-controller was proposed based on the neurons′ excellent ability in information processing to overcome the velocity jump issue in the traditional control method. Then, an optimal sub-controller consisting of a cost function and four constraints was obtained based on the MPC principle. Finally, a terminal region and a terminal sub-controller were designed to stabilize the whole control system. Simulation results with the proposed control method indicate that the converge time of the WMR system to the reference trajectory has reduced from 12 to 5 s and the ranges of the initial linear and angular velocities are narrowed from ［-3, 4］ m/s and ［-5, 6］ rad/s to ［0, 2］ m/s and ［-3, 3］ rad/s, respectively. The output of the system is smooth and bounded, fulfilling global asymptotic stability as well as higher tracking precision. As the algorithm used in derivation is not be limited by the WMR kinematics model, it can be used in other types of mobile robots.

The preparation method and performance test of Carbon Nanometer Tool(CNT) electrodes were explored on the basis of a nanometer test system. Firstly, a CNT electrode was prepared by voltage arc discharge. Then, the volt-ampere characteristic of the CNT electrode was measured online to analyze the resistance variation of CNT and the tungsten needle before and after welding. Finally, the contact characteristic between CNT and tungsten needle was improved by local Joule heating, the resistance of CNT electrode was reduced by Au sputtering, so that the holistic conductive performance of CNT electrode was improved greatly. Experimental results indicate that the resistance of CNT electrode is about 130 kΩ, which is much fewer than the sum resistances of CNT and tungsten needle before welding. The resistance of CNT electrode is reduced to 55 kΩ after 90 s local Joule heating, which is 60% lower than its initial resistance and improves the contact characteristic between CNT and tungsten needle obviously. Moreover, the resistance of CNT electrode is reduced to 40 kΩ after Au sputtering, which is 70% lower than its initial resistance and, improves its conductive performance effectively.

A calibration method for kinematic parameters of a seeker was researched to improve the Line of Sight (LOS )static pointing precision of the seeker. Based on the opto-mechanical structure feature of the seeker stabilized platform , the ideal kinematic model of the seeker was established by local product-of-exponentials (LPOE) formula. Then, the ideal kinematic model was corrected by analyzing the impact of axial errors on kinematic parameters. According to the principle of parameter calibration, a genetic algorithm was designed to improve the searching efficiency by using multi-point crossover and adaptive mutation probability and to obtain the global optimal parameters. By using LOS static pointing error as the objective function, the kinematic parameters of seeker were identified based on the genetic algorithm. Finally, two groups of data were measured by a laser tracker to identify the kinematic parameters and verify the calibration model. The experiments indicate that the LOS static pointing precision can be improved from 109.9″to 22.3″after calibration, which concludes that the kinematic parameters of seeker can be corrected by the proposed calibration method.

Axial magnetic bearing has a lower load capacity and enhanced nonlinear characteristics when there is no current in one of the coils of magnetic bearing. In order to increase the load capacity and to stabilize the rotor successfully, a combined fault-tolerant control strategy was proposed integrated by feedback linearization and guaranteed cost control. First, an axial nonlinear dynamic model of axial magnetic bearing-rotor system was established, and it is linearized in a wide range by feedback linearization method. Then, on the basis of varied parameters, an optimal guaranteed cost controller was designed to control the rotor to be suspended in stability and robustness. Finally, several kinds of experiments were performed on the magnetically suspended rotor when there was no current in one of the coils. Obtained results show that the proposed controller can successfully stabilize the rotor when the bearing loads the gravity of rotor. When the parameter is changed by 35%, the peak value of displacement is 2.6 μm, the overshoot is less than 3%, and regulating time is 82 ms. The results demonstrate the validity of proposed controller and show its excellent fault-tolerant control ability and dynamic and static characteristics.

The modeling methods for thermal errors of a ball screw was researched to improve the positioning precision of a sem-closed loop ball screw driving system. The dynamic characteristics of heat sources, temperature fields of the ball screw were analyzed and how to establish the thermal error model based on Elman neural networks and operating conditions was proposed. Firstly, the internal heat sources and temperature field distribution characteristics of the ball screw were determined according to the structure studied. Then, the dynamic nonlinear functional relationship between the thermal deformation of ball screw and its internal heat sources was researched based on the temperature distribution function. Finally, on the basis of Elman neural networks, the thermal error prediction model of ball screw was established in consideration of the effect of operating condition on the thermal error. Experimental results indicate that the estimated residual error of thermal deformation by the proposed model varies from -3.1 μm to 2.4 μm when the ball screw is worked at a complex environment. It concludes that the prediction precision and robustness by the proposed model based on Elman neural network and operating conditions are better than those by the BP and Elman neural networks (just considering temperature data), and it shows a powerful engineering application prospect.

As the changed temperature of truss structure in a long focus camera will effect the image quality of an optical system, this paper explores the thermal design of supporting truss of the camera. First, the absorbed heat flux and heating powers of a truss rod were analyzed based on the temperature limit requirement of the optical system and the orbital parameters of the satellite. According to the analysis of thermal balance equation, it points out that the temperature difference of the truss is mainly determined by the heating power and arrangement of the heaters. Then, through the theoretical analysis of a simplified mathematical model of the truss, the temperature difference equation of supporting truss was derived, and two thermal control methods including increasing heater areas or increasing the effective thermal conductivity of the truss were presented. A new kind of material named Pyrolytic Graphite Sheet(PGS) was proved to be a candidate that can well satisfy the design requirements of increasing thermal conductivity and lighting the weight of supporting truss. On the calculation,it shows that the temperature difference has decreased from 7.9℃ to 0.83℃ with the mass of each truss increasing only 12 g by using 0.076 mm thick PGS. Finally, through the thermal vacuum test and temperature difference measurement test, it demonstrates that the analysis and simulation is valid and the PGS can withstand vacuum and temperature changes under the test conditions. The method presented in this paper is practical and effective.

As existing sparse coding methods for single-image easily lead to incorrect geometrical structures in reconstructed images, a sparse coding method combining the incoherence constraint of dictionary and the nonlocal self-similarity constraint of sparse coefficient was proposed . Meanwhile, a double layer reconstruction scheme based a smooth layer (SL) and a texture layer (TL) was presented to overcome the over-smooth edges and blurring problem of the reconstructed images because of introducing the nonlocal self-similarity constraint. The method uses a global non-zero gradient constraint SR model to reconstruct a High Resolution Smooth Image (HRSI), and takes the proposed sparse coding method to recover the HR Texture Image (HRTI). Finally, a global-local constraint optimized model were proposed to improve the quality of the final output image. Experiments indicates that the average values of Peak Signal to Noise(PSNR) and the structural similarity (SSIM) have increased 0.798 7 dB-3.242 4 dB and 0.018 6-0.083 5 as compared with those of some recent representative algorithms。The results demonstrate that the method not only improves the subjective vision obviously, enhances the robustness, but also reconstructs more accurate structures and edges, and receives better reconstruct images.

As original compressive tracking algorithms can not select adaptively the object futures, it will result in drifting or tracking lost when the object is occluded or its appearance changes. To address this problem, this paper proposes a real-time compressive tracking algorithm based on online feature selection. First, two complementary projection matrixes were generated in an initial phase, and the projection matrixes were used to extract the feature to construct a feature pool. Then, features with high confidence scores were selected from the feature pool by a confidence evaluation strategy and these discriminating features and their corresponding confidence scores were utilized to construct a Naive Bayesian classifier. Finally, the classifier was taken to process candidate samples by binary classification and response results to the classifier were taken as tracking results. However, the previous result was used to online update the feature pool and the classifier to prepare for subsequent processing. The tracking performance of proposed algorithm is compared with that of original compressive tracking algorithms on several public testing video sequences. The comparison shows that the proposed algorithm improves the tracking accuracy and robustness, and the processing frame rate is 25 frame/s on a 320 pixel×240 pixel video sequence, which meets the requirements of real-time tracking.

An Interactive Multiple Model (IMM) filter algorithm based on Expected-mode Augmentation (EMA) named EMA-IMM algorithm is proposed to overcome the uncertain model and time-varied model parameters of the integrated navigation system for an Autonomous Underwater Vehicle( AUV) in a tough environment. The EMA-IMM algorithm mainly uses the probability of models obtained from the recursive estimate processing for making decision. It filters for the base grids of fixed structure to obtain a fined amendatory model set firstly. Then the amendatory model is filtered to obtain an expected model consisting of a small number of amendatory model grids that are close to the real model. Through a further filtering using the expected model, the suboptimal solution approximate to the real model will be ultimately achieved. Simulation results on the integrated navigation system show that the EMA-IMM algorithm can improve the estimation precisions of longitude and latitude by 97％and 44% respectively as compared with the Kalman filtering algorithm and by 22% and 19% with the IMM algorithm, which proves the superiority of the proposed EMA-IMM algorithm.

The impact of a satellite panel deployment on satellite attitudes was researched and deployment strategies were optimized. A theoretical model for deployment of a single solar panel was established to obtain the analytical solution of the rotation angle changed with time. Combined with the practical application of a satellite project, the ADAMS dynamics of the single solar panel was simulated, and a new ADAMS model approximate to an actual situation and parameter settings were performed. Through deployment test of single solar panel deployment mechanism, the drive moment and resistance moment were obtained to correct the theoretical model and ADAMS model. The correction results show that the difference among theoretical calculation, simulation and test results is within 5%. For a 10.6 kg panel, deployment time has been shorten by 0.4-0.5 s and the impact load augmented about 400 N with each increase of 0.09 N·m of drive moment. The revised ADAMS model can be applied to deployment dynamic analysis of satellite panels and can obtain the effect of different deployment strategies on satellite attitudes. Obtained results provide credible references for selection of deployment patterns and precise controls.

To realize nonlinear encryption of optical images, an optical encryption method based on random Fractional Mellin Transform (FrMT) was established and a corresponding encryption configuration was constructed. In this configuration, two mutually conjugated random phase masks generated by the chaotic map were placed on the input and output planes of an optical Fractional Fourier Transform (FrFT) lens, respectively. The essential role of random phase masks was used to process randomly the kernel function of the FrFT to obtain the random FrFT. The random FrMT was composed of a log-polar transformation and a random FrFT. An optical image was encrypted into one complex-valued ciphertext with the random FrMT, by which the dual encryption of was implemented by the secrecy of pixel value and pixel position simultaneously. The mean square error (MSE) between the decrypted image and the input image for all keys were calculated. It shows that the MSE of the decryption image can be magnified over 200 times when the seed of the chaotic map as the key is increased 10-16, which expands the key space of the encryption algorithm. Moreover, the fractional order of the random FrMT which is taken as a key also has a high sensitivity. Numerical simulation results demonstrate the feasibility and the effectiveness of the proposed method, and the performance analysis for noise addition and occlusion of the encrypted image shows that the algorithm is robust.

On the basis of third-order predictor and backward pixel search technology (IP3-BPS), a lossless compression third-order predictor algorithm using three-stage prediction with adaptive predictor reordering was proposed to overcome the calibration-induced data correlation of hyperspectral images. Firstly, hyperspectral images were divided into groups adaptively according to the correlation factor between adjacent bands. Then using the calibration-induced data correlation and the band scaling factor, a recursive Bidirectional Pixel Search (RBPS) method and an adaptive band grouping method were proposed, respectively, for these groups with spectral correlation factor more than 0.9. The proposed algorithm takes the recursive bidirectional pixel search and the backward pixel search as the last two predictors, and adjusts adaptively their orders to achieve better prediction values. The experiments on the images from an Airborne Visible/Infrared Imaging Spectrometer (AVIRIS 1997) were performed. It shows that the average bit-rate of the proposed algorithm is 3.85 bpp, 0.07-1.28 bpp higher than those of other lossless compression algorithms. It is an effective lossless compression method for hyperspectral images in low computational complexity.

To solve the problem of image reconstruction of incomplete projection data from a short-scan cone-beam CT, a novel cone-beam CT short-scan reconstruction algorithm was proposed based on projection-contraction method.Aiming at the non-monotonic convergence of the Gradient-Projection Barzilari-Borwein (GPBB) algorithm, the predictor-corrector feature of projection-contraction method was analyzed and incorporated into compressed sensing image reconstruction algorithm.The objective function descent direction and the projection onto convex sets descent direction were combined to correct the results of GPBB algorithm to improve the non-monotonic convergence of GPBB algorithm.Then,the experiments were conducted on simulated projection data and phantom scanning data.The simulated results for 25 sampling angles show that the signal-to-noise ratios of images reconstructed by PCBB algorithm are 9.487 0, 9.802 7, 3.615 9 db higher than those of images reconstructed by Adaptive Steepest Descent-Projection onto Convex Sets algorithm, projection contraction algorithm and GPBB algorithm, respectively.The simulation results indicate that when a small amount of projections are acquired, the new algorithm has effectively suppressed strip artifacts and the reconstructed images show clear edges.The algorithm can greatly improved the qualify of images reconstructed from few projection data.

A new image enhancement algorithm based on fractional differential was proposed to improve the lower visibility of traffic video images in bad weather or complicated situations. The characteristics of fractional differential operator (Tiansi) were researched and some points to be ignore in the existing fractional order differential template were applied to the new fractional order differential treatment to improve and update the fractional differential operator. The improved operator was compared with the Tiansi operator and 5 kinds of common image enhancement algorithms. The comparing results show that the proposed algorithm can enhance the details of an image with less noise. When a color image is processed, the improved Tiansi operator can maintain the image colors well , sharpen the grey level images with high frequency signals and do not increase the noise more. When a gray image is processed, the noise can be suppressed by increasing the pixel weight in the template center with much high frequency information. The different fuzzy traffic video images under various situations are tested. The testing results show that the studied algorithm is suitable for the vague traffic video images under bad weather, and the processing speed of the algorithm can meet the requirements of the real applications.

To improve the software reliability of the host computer subsystem in a simulation experimental system, a system-level House of Software Reliability (HOSR) was established. The Run-time Infrastructure(RTI) management module was taken as an example, the dependence of failure modes on the failure causes was analyzed and the relationships between the failure modes and the failure causes were disscussed. The Failure Mode and Effects Analysis (FMEA) and Quality Function Deployment (QFD) were introduced. And then, the HOSR formed by FMEA with the system concept house from the QFD was described. According to the software level diagram of the host computer subsystem, the established library of failure modes and failure causes was used to fill HOSR. Finally, the failure of RTI management module was analyzed on the HOSR analysis matrix. The analytical results indicate that, in RTI management module, there are 23 potential failure modes, 35 potential causes of failure, 116 relationships between failure modes and failure causes, 9 relationships between the failure modes, 14 relationships between the failure causes. analysis method can distinguish failure modes and failure causes much more comprehensive, especially for common-mode failures and common-cause failures. Meanwhile, it can significantly reduce the defect omission of brainstorming and staff workloads.

A novel edge-detached image restoration algorithm with ringing reduction was proposed to suppress ringing artifacts from restored degraded image. Firstly, the support size of a Point Spread Function (PSF) was estimated via an image quality assessment based on a blur identifier, and the strong edges in the degraded image were extracted and their corresponding areas were smoothed to get an Edge-Detached (E-D) blur image. Then, the E-D blur image was restored through the iterative blind restoration using the estimated PSF. Finally, the restored E-D image was fused with the enhanced extracted-edge zone to obtain the integrated deblurring image and to be enhanced further. In order to avoid noise amplification, the Wiener denoising combined with Discrete Wavelet Transform(DWT) denoising were used only in the wavelet low frequency domain of the blur image. Experimental results show that this algorithm can effectively suppress ringing artifacts, reduce the ringing metric of the restored image at least 34% and 12.5% compared with those of the traditional iterative restoration algorithm and the fuzzy filter based algorithm. Moreover, the algorithm improves the recovery image quality evidently.

This paper explores a method to correct the image movement model of an off-axis Three-mirror Anastigmat(TMA) optical system to obtain higher mapping accuracy. Two kinds of line-array Off-axis TMA mapping cameras were modeled and the calculation formulas of image motion speeds and drift angles for the oblique-view camera and the nadir-view camera were deduced by considering earth curvature. By taking the off-axis TMA aerospace mapping camera with an intersection angle of 25° for an example, the influences of image motion speed matching remains and drift angle remains on the image quality of off-axis TMA aerospace mapping camera were analyzed at the oblique-view camera adjusted with the row transfer periods and their drift angles. Results indicate if more 5 TDI stages are used, the oblique-view camera and the nadir-view camera under a constraint of 5% MTF reduction should adjust their row transfer periods respectively; if drift angle of the oblique-view camera is adjusted according to the nadir-view camera, TDI stages should be no more than 71. However, If drift angles are adjusted by giving attention to the oblique-view camera and the nadir-view camera, the imaging quality for the both cameras has no essential influence at a TDI stage of 96.