A new test method based on a Computer-generate Hologram(CGH) was presented for precisely testing off-axis aspheric surfaces and aligning the optics in tests. In this method, off-axis aspheric surfaces were tilt-translated and tested as an on-axis freeform optics to minimize the wavefront compensation and to reduce the relative aperture of the test setup. A software was developed to design the CGH to implement three kinds of diffractive sections,i.e. compensating wavefront in null test, alignmenting the optical path by adjusting the relative position between CGH and interferometer, and projecting fiducial marks around the optics. As an example, a typical off-axis aspheric surface with a figure error less than λ/50 rms was tested in this method. Results indicate that the CGH design error is better than λ/10 000 rms and the total error is better than λ/100 rms.These data agree well with the test result with null lens. The experiment also verifies the feasibility of alignment and fiducial sections and demonstrates that cross-lines projected by CGHs can be used as fiduciary to guide the alignment of the optics and measure the off-axis distance.

A mathematical model for structure light measurement was established based on the inspecting principle of coded structure light, and the optimal system configuration parameters were obtained. The interrelationship between model parameters and their errors was investigated. Based on the system geometry structure, the computing model of depth information was established. By analyzing the influence of system parameters on the pattern imaging, the dynamic deformation of pattern images were captured under different measurement conditions,and the geometry relationship of image coordinate changing with the depth was determined. By analyzing the system parameters and depth information errors, system integration constraint condition was observed,and the optimal system figuration was established. Experimental results show that the system relative error established with optimal parameters is about 0.3%, which can meet the high-precision measurement requirement of 3D reconstruction of coded structure light.

In order to realize the alignment of large aperture telescopes, the RC telescope alignment techniques were investigated based on the image processing and analysis of out-of-focus stellar images. Firstly, the relationship between the disalignment comas due to decentered or tilted errors and the centers of contour circle was analyzed according to transversal aberration. The decrease of ellipse eccentricity with the increase of major axis length caused by residual astigmatism were investigated after removing the coma. Then, the dependence of spherical aberration caused by longitudinal errors on the least circle radius of defocused stellar image was analyzed. Finally, the analysis above was verified by the emulation of Zemax modeling and how to use the out-of-focus stellar image to guide the alignment process and how to process the image were discussed. Experimental results indicate that both the lateral and longitudinal errors are well consistent with the theoretically analyzed results. The precision errors of coma and astigmatism adjusted respectively by tilt and eccentricity are both less than 5%.

A UV-visible spectrometer based on a micro-silicon-slit was developed by using the Micro-electro-mechanical System(MEMS) technology. The generated reasons of the stray-light from the UV-visible spectrometer were analyzed and the relationships between the straightness and thickness of micro-silicon-slit with the stray-light were discussed. Then, an experimental verification was carried out for the stray-light caused by the slit straightness. Furthermore, a new combining filter method was proposed to solve the problem that the radiation power of UV region is much lower than that of visible light region and it is seriously affected by the stray-light when the spectra are measured by a spectrometer with the halogen tungsten light source. The method combines a neutral density filter and a balanced filter in the front of the micro-silicon-slit and the detector to balance the radiation powers of different spectral regions and to inhibit the stray-light of UV spectrometer. It is shown that the stray-light of the UV-visible spectrometer has been reduced by 23% as compared with that by using the traditional method.

For the special structure of cylindrical inner surface of a Wolter-I grazing mirror, the removal function of an elastic ball tool with deflected rotation motion was researched,and the relations between the max removal depth of removal function, dwell time and other parameters obtained by experiments were demonstrated.By using a 2-step method for polishing, it shows that when the ratio of rotating angle velocity to tool-workpiece is 1.41, the surface quality is in the best state. Furthermore, the fabrication of Wolter-I grazing mirror base on Computer Control Optical Surface(CCOS) was introduced. In the research, Zerodur glass was selected as mirror substrates and it was polished by CeO abrasive on a Digital Control(NC) platform developed by ourselves. By changing the distance between the tool and the workpiece, the removal functions in different contact areas were determined in the experiments, and the residual error from previous polishing process was corrected effectively. Using two kinds of measurement devices to outline the surface profile, it shows that the final surface figure of the mirror is 1.39 μm (PV), 0.34 μm (RMS) and the error of circularity is less than 0.1 μm (RMS). These results demonstrate the feasibility of fabricating Wolter-I grazing mirror with the 2-step deflected rotation method.

In order to improve the efficiency of ultrasonic bonding process for a polymer Electro-Mechanical System(MEMS) and to prevent the flow of polymer under a high temperature,a micro energy director array was proposed. Based on the experiment of PMMA micro-pipe sealing, the influence of micro energy director array on the process of ultrasonic sealing was studied. Five kinds of micro energy directors with different heights were fabricated on the sealing surface by a hot embossing technique. The method of ultrasonic precise sealing based on acoustic transfer efficiency feedback was used in the experiment to study the influence of the height of the micro-array structure on the quality of ultrasonic sealing and to observe the wetting behavior of micro-array structure. Experimental results indicate that the ultrasonic energy to achieve complete connection increases with increasing the height of micro energy director array. The micro energy director can control the flow of the melting-polymer effectively and can obtain the uniform sealing surface and high quality precise sealing.

With the aim to improve the performance of an infrared detector and to enhance the characteristics of a Focal Plane Array(FPA), a capacitive-read infrared FPA with the structure of cantilever element in a fold-line leg and a long-line leg was designed. The thermal conductivity , thermo-mechanical sensitivity and the thermal time constant of the structure were discussed and optimized, then these parameters were set to be 3×10-7 W/K, 162 nm/K and 35 ms, respectively. The sensitivity of design is fast enough to satisfy the requirements of human eyes.Furthermore, the Noise Equivalent Temperature Difference( NETD) which is key for the FPA detection was discussed, and calculated result shows that the NETD of the designed FPA is 6 mK. A simulation for the NETD was also performed, which shows the NETD is 5 mK. Experiments indicate that the simulation result agrees with that of the calculation well and show that the NETD is small enough to satisfy the resolution for signal read-out. Finally, the capacitive-read FPA was fabricated based on silicon micromachining technology successfully.

Fracture splitting notches from C70S6 steel used in a connecting rod and from gray cast iron used in an engine crankcase main bearing bracket were processed by a YAG pulse laser, and the fracture splitting experiment was conducted in the fracture splitting device. The geometry appearances, microstructures, micro-hardness change as well as the fracture surface of fracture splitting notches were studied by a tool microscope,a scanning electronic microscope and a micro-hardness tester. Characteristics of fracture splitting notches processed by the pulse laser were discussed, and the influence of pulse laser processing fracture splitting notches from ductile/brittle materials on the expansion and breaking performance of the notchs was revealed. The results indicate that the geometry appearances and notch effects of the fracture splitting notches from ductile/brittle materials processed by the pulse laser are obviously superior to those of the broaching and linear cutting processing, and the gap sensitivity of gray cast iron has been strengthened greatly. The brittle martensite has been produced nearby the fracture splitting notches from ductile/brittle materials, and micro crack appears in the root of notch. The organization is obviously refined. The micro-hardnesses of melting zone and the solid phase transition zone of ductile/brittle materials approximately are 820/850 HV and 860/550 HV,respectively,which are higher than that of the substrate, and show a remarkable hardening effect. The ductile tearing phenomenon can not be seen in the C70S6 steel notch root zone. The ductile/brittle material fracture shows the brittle fracture characteristic,which satisfies the fracture splitting technological requirement.

The influence of vehicle platform deformation on the measuring errors of a theodolite was analyzed,then the platform deformation was divided into two types of translation and rotation. A numerical simulation was performed to demonstrate that the platform rotation deformation was the main factors to affect the measuring error of the platform.A simulation model of vehicle theodolite between measuring error and rotation deformation was established,and the influence of the target location and rotation deformation on measuring errors was simulated based on the model. Furthermore,the platform angular deviation was measured by a self-collimation measurement system based on Moiré fringe and the platform angular rotation was measured by a inclinometer, which then was validated by experiments.Obtained data show that this method can effectively compensate static angle measuring errors come from the deformation of vehicle platform, and can improve the azimuth precision to 103.7″and elevation precision to 89.4″. It provides a theoretical reference and technical support for the high-precision measurement of vehicle theodolites.

A nonlinear mechanical model for sub-micron cantilevers was analyzed in detail and its physical mechanism was researched to provide a theoretical basis for experiments.The electrostatical force was used to achieve the resonance state cantilever beams and a Polytec laser Doppler vibration measurement system was taken to observe the frequency response curve. Experimental results show that the beams have significant nonlinear effects (the spring softening effect) and the nonlinearity is relatively independent on the AC voltage, but it is markedly enhanced with increasing the DC voltage. The obtained maximum peak shift is 0.5 MHz and extracted first-order mechanical elasticity coefficients are 79.62, 31.75, and 14.92 N/m, respectively. Furthermore,the deviation of the experiment was also disccussed and analyzed.The effects of overetching by wet chemical etching on the stiffness and frequency response were stimulated by ANSYS software.In conclusion, the corresponding simulation results are well coincident with the experimental data.

A designing structure which can harvest multi-direction vibration energies was proposed by using the new Rainbow shape piezoelectric energy transferring elements.In order to analysis and test the electricity generation of the piezoelectric energy transferring elements,the finite element analysis and practical tests were performed.The simulation results indicate that the open circuit voltage of Rainbow shape piezoelectric energy transferring elements will be reduced with the increases of the width and thickness of a metal substrate,the width and length of a piezoelectric film and the initial curvature radii of energy transferring elements; and it will be increased with the increase of the length of the metal substrate and shows a maximum with the increase of the thickness of piezoelectric film.The experimental result demonstrates the validity of finite element analysis and emphasizes that the open circuit voltage of energy transferring elements will have a maximum when the thickness of piezoelectric film is 0.25 mm.Furthermore,in the output power tests,the maximal output power of Rainbow shape piezoelectric energy transferring elements is about 7.75 μW. The results of simulation and experiment provide helps for designing, manufacturing and realizing the Rainbow shape piezoelectric energy transferring elements.

A scheme for the motion compensation of imaging spectrometers by using a scanning mirror drive system consisting of a step motor,a harmonic reducer and a obsolute encoder was introduced.The rationality of selected five-phase hybrid stepping motor in the system was illustrated. On the particularity of the five-phase hybrid stepping motor, its drive technology and control methods were studied. Firstly, the special requirements of the scanning mirror and the drive method for step motor were pointed out. Then, for the two operating conditions which scanning mirror ran in low-speed and in high-speed to the initial position, a new drive method based on the new pentagon winding connection was proposed. Furthermore, the influence of the drive method on the enhancing performation of scanning system was analyzed. Finally, the related equipment and appropriate measuring methods were used to achieve the precision of scanning mirror motion. Obtained results show that the Root Mean Square Error(RMSE) of angular velcity is less than 5.5% in scanning mirror at low speed of 0.1~0.5 (°)/s in uniform speed motion and given tracking compensation curves. The imaging spectrometer can basically meets the requirements of scanning applications.

A micromechanical Silicon Resonant Accelerometer (SRA) prototype fabricated by Deep Dry Silicon on Glass (DDSOG) process was presented. The amplification of single-stage microlever in the SRA to the interial force was researched,then it pointed out that the scale factor of SRA is proportional to the system amplification factor n.The theoretical model of single-stage microlever in SRA was established, and the formulas for microlever amplification factor A and axial stiffness K were derived. On this base, the formula for system amplification factor n was also derived. According to the formula, the theoretical value of system amplification factor n for SRA prototype is calculated to be 21.820. As a comparison,its simulation value of 19 is also achieved by the Finite Element Analysis (FEA). The practical test for the SRA prototype was performed,which shows that the scale factor of the SRA is 143 Hz/g and the system amplification factor is 25.466. As compared with the theoretical value,the errors of simulation and test results are 14.8% and 14.3%,respectively, which are within the accepta-ble range.

A Linear Auto Disturbance Rejection Controller(LADRC) is designed to eliminate the disturbance from the variational aircraft gesture angle velocity on forward image motion compensation of the reflector in an aerial camera. The research includes the equivalent relation between angle velocity and torque disturbance, as well as the estimation and compensation methods of disturbance. Firstly, the influence of angle acceleration on the image motion compensation is analyzed. Then, a mathematic model of reflector system is established with consideration of the torque disturbance, and a linear extended state observer to estimate the disturbance and a control law with disturbance compensation link are designed. Finally, an experiment on the reflector is performed to test the control performance of the LADRC. Experiment results indicate that residual error of LADRC is within 1.4%, which has reduced 75% as compared with that of traditional lag-lead compensation method.Furthermore,the stable state speed precision of the LADRC reaches 0.25%. In conclution,the control method improves the reflector performance of cameras in speed precision and robustness.

To solve the low efficiency of oxygen mass transport and weak performance of self-breathing micro-direct methanol fuel cells (μDMFCs), the different cathode current collectors with a perforated structure were designed. Firstly, the effects of opening shapes and opening ratios on the cell performance were studied by establishing the cathode model of μDMFC,which indicates that the opening shapes have little effect on the cathode current density, and the variations of cathode current are rather small in the range of opening ratios. Then, experiments were performed to verify the simulations. Based on the above analysis, a cathode perforated structure with parallel channels was presented. Compared with conventional structures, the simulation results illustrate that the improved structure can effectively improve the oxygen mass transport to increase the cell performance. A self-breathing μDMFC with the active area of 8 mm×8 mm was fabricated on the stainless steel plates by utilizing micromachining technology. Test results show that the peak power density of the μDMFC is 11 mW/cm2 with methanol solution of 1 mol/L and speed of 1 ml/min at room temperature. In conclusion, the improved self-breathing cathode structure can be contribute to further development of portable micro power source systems.

A combined profilometer combined atomic force probe scanning measurement with non-contact optical measurement based on one interference microscope substrate was developed for ultra-precision surface topography. The white light interference microscope based atomic force probe scanning measurement, the relationship between the deflection of an atomic force probe cantilever and the movement amount of interference fringes and the nonlinearity error correction method of probe cantilever deflection measurement were given.Then, the contact operation mode combined the displacement of vertical scanning system and the deflection of the cantilever was analyzed.A 3D precision displacement system and a AFM head based on white light interference microscope were developed. Under the same condition, the calibration grating TGZ2~~PTB(107±2nm) made by NT-MDT Co., was measured 10 times on the same region. The measurement results show that the standard deviation is 0.96 nm and the relative repeatability error is 3.08%. Several examples measured by atomic force probe scanning method, phase-shifting interferometry and vertical scanning white light interference method were given, respectively.It concludes that the combined profilometer can be used to measure ultra-precision machining engineering surfaces, optical surfaces,and micro-nano geometric structures.

In order to expand the application scope of Atomic Force Microscopes(AFMs), a large-range and high-speed AFM was developed. A feed-forward and feed-back algorithm is proposed to deal with the Z-control of the AFM. The feed-forward controller includes an auto leveling controller and a last line based controller. The former determines the location of sample tilt by multi-lines scan, and the tilt displacements of all scan points are expressed by a formula.Then,it is transformed into Z control voltages to drive the down-stage. The latter uses the last line height signal as the reference input of Z-control in the current line scan. The feed-back controller is a proportional integral(PI) controller whose P parameter is dynamically changed with error signals. Experimental results indicate that the max error value by proposed method is decreased from 40.17 nm to 6.01 nm,and the root-mean-square error is decreased from 22.85 nm to 2.01 nm. It suppresses error signals evidently, enhances the ef-fectiveness of Z-control,and makes image more precise.

Dynamic speckle correlation method was presented to measure the piezoelectric ceramic characteristic curves and scale linear intervals in this paper. Firstly, the speckle intensity cross-correlation of a projection phase diffuser in the micro-system was calculated, and influences of the resolution and magnification of the micro-system on object displacement measuring were also discussed. During the calculation of cross-correlation, a progressive correlation method was used to avoid the de-correlation caused by the object surface. A sampling system for micro speckles was designed and the hysteresis curve of displacement induced by a piezoelectric ceramic was measured. Experimental results indicate that the theory precision of the measuring system is 0.082 μm when a micro objective is with magnification of 100 and NA of 1.25. In consideration of the diffraction limit, the theory precision of the measuring system is 0.348 μm. It concludes that the method satisfies the requirements of measuring and scaling of linear intervals. Comparing with other methods,the dynamic speckle correlation can improve the calculation speed and measuring precision.

The compressive and tensive characteristics of Magnetorheological(MR) fluid were investigated with a magnetic field. A experiment setup was designed and fabricated to test the compressive and tensive properties and the magnetic behavior of the equipment was analysed by using the ANSYS/Multiphysics. The compressive and tensive resistance of the MR fluid was then measured for different field strengths, and the relation between tensile yield stress and shear yield stress for the same magnetic field was developed. The compressing tests show that the MR fluid has a smaller compressive modulus when a compressive strain is about 0.15. The compressive stress and compressive modulus have an exponential relationship with the compressive strain when they are higher than 0.15.Moreover, the exponent increases with the enhancement of the applied magnetic field. The tensile yield stress is about four times of shear yield stress. According to the results above,The shear yield angle is calculated to be 13.8-16.9°. Obtained results prove the retionability of the physical model for describing the shere stress of MR fluid.

In order to improve the image quality of the space TDI CCD remote sensing cameras, the method to set optimally of TDI CCD integration stages and gains was proposed. The dependence of the Signal to Noise Ratio(SNR) and Modulation Transfer Function(MTF) on the integnation stages was discussed.Then, it points out that increasing integration stages can resolve the lack of exposure light energy and improve SNR,but it causes the image MTF reduced. However,increasing gains can also resolve the problem of lack of exposure light,and it makes the SNR and the imaging MTF unchanged. These results means that the set of the SNR and imaging MTF can improve the image quality. By using the SNR×MTF as the optimal goal, this paper optimizes the integration stage and the gain. Numerical calculation result shows that when the pitch speed of satellite is 0.005°/s and the exposure is 1/66 of saturation,the high quality image can be obtained in the integration stage to be 44 and the gain to be 1.5.

In order to robustly track the multi-degree-of-freedom moving objects in video sequences at a complex background, a tracking algorithm for multi-degree-of-freedom moving objects was proposed based on the particle filter principle. Firstly, the similarity of a target model and a candidate model was taken as the structural basis of observation by using mean shift algorithm. Then, based on the kernel-color histogram, the center position of the object and the covariance matrix that described the shape of the object were updated to adjust kernel-bandwidth and modify the size of tracking window,then to implement the tracking for multi-degree-of-freedom moving objects. In particle filter, the number of particles is to be 100, the variance of coordinate components is 5 in the covariance matrix,and the variance of scale and angle components is 0.1. Tracking experiments for various objects in different scenarios show that the proposed algorithm can track multi-degree-of-freedom moving objects steadily, and can adapt to the change of scales and angles for objects.

On the basis of angle measurement by a rotating planar laser in the single station and intersection in the multi-station, a coordinate measuring system was introduced，and its calibration method was optimized. With analyzing its features of special angle measurement and multi-sensor intersection measurement, the principle of azimuth information measurement by the single station and the system structure were described,then the calibration methods for structure parameters of single station and system parameters were studied. According to the characteristics of a rotating shaft and a laser plane, structure parameters were calibrated by theodolites through designed corresponding accessories,and the system calibration was realized by determining the laser planes in the initial place by using a receiver. In order to perfect the calibration technology, the main error factors affecting the calibration accuracy were analyzed, the calibration methods of structure parameters and the process of system calibration were modified.An integrative calibration method for the system was proposed. Experiments show that the measuring accuracy of the system has reached 0.1 mm,which proves the calibration method is feasible, and can improve the calibration accuracy.

For hyperspectral image classification, a Support Vector Machine (SVM) algorithm with composite kernels was presented to fuse both the spectral information and spatial information of the image. The algorithm adopts Principal Component Analysis (PCA) algorithm to extract the image feature and reduce the dimension for hyperspectral image,and uses the Virtual Dimension (VD) algorithm to estimate the Intrinsic Dimension (ID) of the image. Then, the remained number of Principal Components (PCs) was determined on the basis of the ID.Furthermore, spatial features were extracted by mathematical morphology from the remained PCs,and the Extended Morphological Profile (EMP) vector of image was obtained. By combination of different strategies to construct composite kernels, the spatial information was introduced into the classifier to implement the classification with the SVM and based on both the spectral information and spatial information. Hyperspectral image experiments indicate that the overall accuracy and Kappa coefficients of the proposed approach increase about 2% without increasing the training time obviously. Compared with the classifiers only using the spatial or spectral information, the proposed method shows a lot advantages.

To improve the recognition ability of a pyroelectric infrared (PIR) detector for different infrared radiation sources, a method for human and non-human recognition based on Canonical Correlation Analysis (CCA) was proposed. Firstly, the frequency spectrum and wavelet packet entropy were extracted as features,and the spectrum was divided into sub-patterns. Then, each sub-pattern and wavelet packet entropy were fused with CCA method, and the fused feature was employed as classification information. By this way, the feature fusion was realized and the redundant information among the features was also eliminated. Finally, the recognition results were obtained by a majority voting method. As a special case of the sub-pattern fusion, the classification abilities of the features fused with their own sub-pattern were also studied in the paper. Experimental results show when the frequency is divided into 5 sub-patterns, the recognition rate can reach 95.2%, which is higher 2.7% than that of only fusing the frequency and the wavelet packet entropy.Moreover, the recognition rate of wavelet packet entropy fused with its own sub-pattern is 90.7% , which is higher 2.3% than that of wavelet packet entropy.

For the difficulty of moving object detection by a moved camera, a method for detecting dynamic background caused by camera motion was proposed by combining a Scale Invariant Feature Transform(SIFT) and a differential multiplication. Firstly, the image registration based on SIFT features was applied to calculate transformation parameters, including translation factors, rotation angles and scaling coefficients,and to provide a robust matching for realizing the motion compensation precisely. Then,the object segmentation based on the differential multiplication was introduced to detect moving objects accurately. Finally, the experiments on the video sequences were performed to verify the robustness and validity of the method.Experiments show that instead of 6 or more frames, 4 frames are enough for the differential multiplication in practical applications. The detection method is robust and denoising in dynamic scenes and has good adaptability to changing illumination, occlusion and camera movement.

With the need of 3D animation design for the 3D facial expression data dynamically, a 3D facial experession acquiring system was designed.Firstly, a new model was constructed by using a feature grating pattern and a space mapping method. Then，a feature grating pattern was designed, and the extraction method of a feature rectangle was provided.Furthermore, the relative phase was translated into an absolute phase by using the information of feature rectangle, and the three-dimensional world coordinate information was obtained by the space mapping method. Finally,the tests on accuracy verification and facial measurement were performed.A standard hemispheroid with (220±0.01) mm in diameter was measured five times，and the results show that the average of the fitting diameter is 219.95 mm, mean square error is 0.48 mm, and the mean square error of the distance between 3D points and spherical surface arrange is 0.28-0.46 mm. Moreover, the face with different expressions of a student was measured by the system,and the results indicate that the 3D data can be acquired by the system with only one fringe pattern and a vivid profile of the face is shown.

The method to detect small moving targets in infrared image sequences that contain moving nuisance objects and background noises is analyzed in this paper. An infrared small target detection algorithm combined with temporal and spatial domains is put forward. On the basis of the background change slowly, the algorithm firstly confirms the object region based on the third central moments of frame difference, and decomposes the frame difference image by nonsubsampled Contourlet transform to define the energy coefficients of the sub-band images. Then the image based on the energy value of each pixel is obtained. Finally, the final detecting of the target is realised according to the different features of small targets, backgrounds and noises. The results indicate that the small infrared target detection based on nonsubsampled Contourlet transform can precisely detect the small infrared target and has better target detection performance. When the small targets with invariable and moving backgrounds are detected, the detection rate of the proposed algorithm can reach 98% and 97% and the mean of false alarm points is only 0.05 and 0.17, respectively.It concludes that the proposed algorithm can detect the small target when the target has fast or anomalistic movement.

In order to reduce the computational complexity and pair the parameters automatically in multiparameter estimation for array sensors, an algorithm for joint the Direction-of-arrival (DOA) signal and frequency signal based on a state-space model was presented. Firstly, a special state-space model with system matrices including the DOA and frequency information was constructed. Then the second order statistical properties were taken to restrain the noise by using appropriate auxiliary variable,and the extended observability matrix was estimated by using orthogonal projection theory and singular value decomposition. Finally, the estimation of system matrices was obtained from the generalized observability matrix,and the DOA and frequency of the signal were estimated from the eigenvalue decomposition of system matrix. Simulation results are presented to demonstrate the effectiveness of the algorithm,and the results show that the RMSE of frequency estimation is 0.003 5 rad and that of DOA estimation is 0.38° under the condition of estimating two signal sources with 0 dB for SNR.

A hardware implementation method for lossless image compression is proposed to overcome the difficulties of embedded wavelet coding methods in hardware implementation and high costs. Firstly, the algorithm divides wavelet coefficients into a low frequency block and three high frequency blocks according to sub-band properties, and then uses different methods to code respectively. In the low frequency block coding method, the Difference Pulse coding Modulation(DPCM) is firstly used to reduce coefficients' dynamic range. Then, a modified bit plane coding method is used to output the bit stream. In the high frequency block coding method, the proposed modified Set Partitioning in Hierarchical trees(MSPIHT) algorithm is used to code three high frequency blocks respectively with their thresholds.The MSPIHT optimizes the outputted bit stream by using a type of A set judge, reduces memory requirement and avoids memory dynamic management by eliminating the lists of SPIHT algorithm.Moreover,the MSPIHT avoids repeated calculation in scanning process and enhances the coding efficiency by adopting MMVS. Experiment results show that the bit-rates of all international standard testing images have reduced more than 1.4 bit/pixel and the coding speed has increased more than three times as compared with that SPHIT. It is concludes that the proposed algorithm is super in real-time performance, low memory requirement and fit for hardware implementation.

In order to resolve the problem that LED display panel may appear position distortion and cause the high nonuniformity of LED display images, an improved method called distortion correction was proposed on the basis of an existing correction technique namely the correction technique based on CCD.The correction technique based on CCD was introduced,then the reason that this technique needs to be improved was expounded. By taking a LED display panel which is spliced by modules for the example, the distortion correction model was constructed. This model modifies the gray value of every LED display panel’s pixel to reduce the nonuniformity according to the nonlinear relationship between distortion and ideal module position, Furthermore, realization steps were described in accordance with the constructed model. Finally, the improved method was implemented in a LED display panel with a resolution of 128 pixel×128 pixel, pixel center spacing in module of 7.62 mm, and the module containing only one pixel. Experimental results show that this method is able to reduce the nonuniformity of LED display images from 8.9% obtained by the correction technique based on CCD to 0.97%.

In order to determine the position parameters between an imaging laser radar and a camera and to establish the mapping relation between them, an improved calibration method was proposed based on Singular Value Decomposition (SVD) and present calibration arithmetics. Firstly, the calibration principle was introduced, then the SVD was used to solve overdetermined equations to obtain the projective transformation matrixes of laser radar coordinates and camera coordinates. To reduce the influence of noise, the initial result was optimized with the Levenberg-Marquardt algorithm. Then, the calibration experiments were carried out with a 3D calibration object and the image fusion experiment was performed according to the calibration result to validate the improved calibration arithmetic. Experimental results indicate that the average calibration precision of the arithmetic is 13.33 μm and the standard error is 7.49 μm,which is doubled that of the original arithmetic. The precision of the improved calibration arithmetic can meet the requirement of image fusion.