The polarization calibration of an airborne multi-angle polarimetric radiometer was researched to eliminate the polarization effects induced by itself and to improve its measurement accuracy. First, instrument detection equations with calibration coefficients were derived according to the main factors that have been induced into the instrumental polarization effects, and the expression of calibration coefficients was solved based on the detection equations for the same signal when the instrument was fixed at states of 0°and 90°.Then, the calibration method to get the signals of the two instrument states and to solve the calibration coefficients were designed. Finally, the polarization calibration coefficients were computed from the measured data using a non-polarized light source and a completely linearly polarized one, respectively, and the polarized calibration results were verified by adjusting the polarization degree of the incident light. The results indicate that the instrument can offer a polarized measurement accuracy better than 0.5%, which meets the requirements of the instrument polarimetry.

As the integral imaging three-dimensional(3D) display technology is limited by the depth of field in a reconstructed image, it can not be widely used in the fields of 3D display. Therefore, this paper proposes a method to enhance the depth of field in a projection-type integral imaging system based on the angle control. To enhance the depth of field, an additional lens is used to control the diverging angle of every pixel in the elemental image array. Contrast to the conventional method with multi-layers, the proposed method has advantages in simpler systems, higher efficiency and lower requests for display devices. Experimental results show that the depth of field is 6.4 times that of original one when the diverging angle is 1.79°. And as compared to the conventional method with an enhancement of 4 times depth of field, the proposed method has increased by 60% in efficiency. These results demonstrate that the proposed method gives a new way to enhance the image quality for integral imaging display systems.

The diurnal variation characteristics of Aerosol Optical Thickness(AOT) in different city regions and seasons were analyzed by using the long-term measurement data from a sun-photometer DTF in several typical regions of China. It indicates that the AOTs in Chinese different regions show complex and different diurnal variation characteristics and the diurnal variation trend is obviously different. The AOT variations of yearly-mean, monthly-mean, and seasonal-mean in each region were also presented. The results show that the AOT in Lijiang of Yunnan Province is the least, where the atmosphere is clean, and the small particles are dominant, and that in Zhangbei of Hebei Province is the second least one. The AOTs both in Kashi of Xinjiang province and Hefei of Anhui province are larger, and the bigger particles are dominant in Kashi while the smaller particles are dominant in Hefei. In every region, the AOT is larger in spring, and the smallest in autumn. Moreover, the value of AOT is between 0.15 and 0.7 in Kashi, 0.08 and 0.4 in Zhangbei, 0.2 and 0.75 in Hefei, 0.01 and 0.1 in Lijiang, respectively. The frequency distributions of AOT and ngstrm index are basically Gauss distribution. The peak value distribution of AOT from high to low is Hefei, Kashi, Zhangbei, and Lijiang in turn and that of the ngstrm index from high to low is Lijiang, Hefei, Zhangbei, and Kashi in turn.

The effect of temperature on the transfer function and space intersection angles of a three-line mapping camera was analyzed and the thermal control index of the three-line mapping camera was calculated by a thermal-optical method. First, the temperature field and thermal elastic deformation were analyzed with Finite Element Method(FEM) on the basis of hypothetic thermal loads and the wave surface fitted with a Zernike polynomial was put into an optical software to explore the effect of temperature on the optical transfer function. Then, the temperature regressive data of transfer function for the mapping camera were obtained in an assumed temperature field. Furthermore, the thermal size stability of a mapping base was analyzed and the effect of temperature on the intersection angles of the three-line mapping camera was gotten. Experiments show that the over or short thermal control designs can be avoided with the proposed method and it can provide reasonable data for thermal control indexes.

An on-line 3D detection system for optic component surface measurement was designed based on digital holography. An off-axis structure was chosen to evade the optical axis of a test object and the tilt phase compensation was used to eliminate the tilted phase distortion caused by an off-axis optic path. In this system, the illuminating angle and view direction could be changed by rotating the object around the optical axis, by which the recording distance was increased and the resolution was improved using synthetic aperture technique. Meanwhile, the superposition of reconstruction images could decrease the effect of speckle noise on the reconstructed distribution. The proposed method is experimentally validated by a resolution test target and an optic mirror, and it shows that the 2D resolution of the system has been 10 μm at a recording distance of 40 cm and the system noise also be suppressed effectively.

The quantitative analysis of spectral domain optical coherence tomography was researched to acquiring interior information of biological tissues.The single-scattering model, confocal single-scattering model and multiple-scattering model were presented and discussed. An average A-scan algorithm and a nonlinear curve-fitting method were employed to explore the scattering features of IntralipidTM. Experiments show that the relation of scattering coefficients and concentrations is linear in concentrations from 1% to 10% of IntralipidTM.Fourthermore,it confirm that the confocal single-scattering model is suitable to our spectral domain optical coherence tomography. In addition, the confocal model was used to study fresh rat liver with a scattering coefficient λ0 of 8.9 mm-1 at 1 550 nm. Obtained results demonstrate that the quantitative spectral domain optical coherence tomography can be used for clinical diagnosis of diseases in a great potential.

To research the uniform axial tension effect of an Electroless plating Ni-P coated Fiber Bragg Grating (ENFBG), the stress and strain properties of the ENFBG were analyzed based on its axial symmetry.The finite element method was used for theoretical calculations, then it was verified by following experiments. Theoretical calculation shows that under uniform axial tension, the core of the ENFBG is drawn uniformly, the shearing strain is far less than normal strain and can be omitted.The normal strain is proportional to the axial tension,and the change of center wavelength of ENFBG is proportional to that of the tension too. However, because of the difference of mechanic parameters between electroless plating Ni-P coating and optical fiber, the metal coating reduces the uniform axial tension sensitivity of the FBG. In sensor experiment, when the thickness of the coating is 7.25 μm, the sensitivity is 12.45 pm/MPa, R-square is 0.999 6,and the theoretical calculation value is 12.744 pm/MPa. Moreover, the sensitivity is decreased with the increase of the thickness of metal coating. The ENFBG is an excellent uniform axial tension sensor,because its center wavelength shows high linearity to uniform axial tension and its metal coating can provide good protection.

On the basis of error propagation characteristics of an Inertial Navigation System(INS) and the effect of axial Ring Laser Gyroscope(RLG) on the errors of longitude and latitude, a calibrating method for the RLG drift was proposed to solve the problems that single-axis rotation could compensate the vertical RLG errors automatically, but could not compensate the axial RLG errors in a single-axis rotation INS. First, the auto-compensation principle of the single-axis rotation INS was introduced briefly. Then, the effect of the axial RLG drift, initial heading and attitude error, and initial velocity error on the longitude and the latitude was analyzed on the stationary base. Depending on the errors of longitude and latitude, a new method based on the least squares was proposed to calibrate the RLD drift precisely. Finally, the static and vehicle navigation experiments were performed. The results show that the identification precision reaches 0.000 5 (°)/h and radial position error of the INS is less than 1 nm/72 h. The axial RLG drift can be calibrated precisely by the least square method proposed and the experiment result of navigation meets practical demands.

As the deviation between the real parameters and the thermal design parameters of a space spectral imaging apparatus in orbit effects on the design accuracy of complete appliances, this paper analyzes the thermal design based on the system sensitivity theory, and establishes the heat balance equations in orbit. Under the design variable analysis of the heat balance equations, thermal design parameters effecting the temperature distributions of complete appliances are given. As an example, the sensitivity of thermal design parameters mentioned above is analyzed for the space spectral imaging apparatus. The analytical results show that the sensitivity of solar absorption coefficient to the mean temperature is equal to zero, and the sensitivities of emissivity , inner heat source, thermal conductivity, and contact thermal conductivity to the mean temperature are 2.2-14.55 ℃, 1.8-2 ℃/W, 2.25×10-3- 4.39×10-2 m℃2/W and 0-1.1×10-3 m2℃2/W, respectively. Test results prove that the sensitivity analysis based proposed thermal design scheme is effective and feasible proved by thermal test results.

A method to estimate the efficiency of relative intensity noise subtraction for a Fiber Optic Gyroscope (FOG) was proposed, and the estimation result could be used to improve the reliability of relative intensity noise subtraction of the FOG. Theoretical analysis shows that the efficiency of relative intensity noise subtraction of the FOG is highly dependent on the cross-correlation between the outputs of FOG and coupler. By proposed method,the experiment on a high precision interferometric FOG was performed, and obtained results show that when the cross-correlation coefficient is 0.91, the variance of the noise after subtraction reduces to 17.16%.However when the cross-correlation coefficient is 0.28, that of the noise after subtraction goes up to 143.18%. These results are coincident with that of theoretical analysis.By utilizing this method,the cross-correlation between the outputs of FOG and coupler improves the reliability of the intensity noise subtraction.

This paper analyzes the measuring results of a diffraction efficiency measuring instrument for concave gratings to improve the measuring accuracy of relative diffractive efficiency. According to the testing theory of diffraction efficiency of concave gratings, the effects of the spectral broadening of exit beam for the concave grating, section changing of diffraction beam, control of light source radiation luminance and the synchronization precision of testing wavelength on the measuring accuracy were analyzed, and several kinds of necessary equations were given. Based on regression analysis and a large number of experiment data, an optimization equation was established and programmed into a testing programming to correct automatically the test results. Experiments indicate that the tested results are correct and more precise, and the difference between optimization value and theoretical one is less than ±2.5%, which improves the testing precision effectively. This method is simpler operation, not to change the components of the instrument, and can satisfy the test requirements in strong real-time and higher precision.

To evaluate the static performance and design optimally a parallel mechanical leg for the six-legged robot, a static analysis method based on the mapping of both constraint and actuation forces was proposed,and the mechanical leg was designed based on this method. Firstly, the constraint and actuation Jacobian matrixes were established based on the mapping of both constraint and actuation. Then, according to the virtual work principle, the actuation statics transmission equation was established, the performance evaluation indexes of actuation statics were designed,and the relationship curve between the performance evaluation index and structure parameters was calculated. By the same way, the transmission equation and performance evaluation index of constraint statics were obtained,and the relationship curve between the performance evaluation index and structure parameters was also calculated. Finally, based on the performance evaluation indexes both of constraint and actuation statics, the structure parameters were optimally designed by Monte Carlo method. Calculations show that when the structure parameters of a fixed platform and a movement platform are 200 mm,and 80 mm and the UPU branched chain lengths l1min,l1max are 500 mm and 900 mm, the statics comprehensive performance of the parallel mechanical leg is the best. These studies lays the theoretical foundation for further study of the six-legged robot.

To estimate the thermal distortion of a large aperture SiC lightweight primary mirror exactly, the calibration and calculation methods for the SiC lightweight mirror were researched in different thermal modes. Temperature sensors stuck on the lightweight mirror were divided into three levels along an optic axial direction, and the temperatures at each level were fitted by quasi-Zernike polynomials, respectively. By taking a 4 m SiC lightweight mirror for an example,the Finite Element Method(FEM) built by MSC.Patran was used to calibrate and calculate the thermal distortion in eighteen kind temperature fields of the first ninth quasi-Zernike polynomials and the Max deformation, surface error PV and RMS values due to the lightweight mirror thermal distortion were obtained for different temperature patterns. Calculation results show that the deformation of lightweight mirror is max under the first quasi-Zernike polynomial temperature pattern, and the surface error RMS value is 278.3 nm. Using a least-square method, the aberration coefficients of surface errors are calculated by quasi-Zernike polynomial wavefront fitting program for each temperature field, and the main aberration caused by the thermal distortion is piston, tilt, defocus, coma and astigmatism.

In the UV-LIGA fabrication process of a microinjection metal mold,the large internal stress of a SU-8 layer can result in SU-8 mould defects such as micro cracks, distortion or even exfoliation. In order to avoid these defects, the ultrasonic stress relief technology was introduced into the microinjection process. First, the SU-8 mould was obtained by coating, soft baking, ultraviolet exposure and post-exposure baking,and it was treated by the ultrasonic stress relief equipment before developing. Then the backless plate growing method was adopted to directly fabricate the nickel micro-electroforming pattern on a 38CrNiMnMo mould steel substrate and the SU-8 layer defects emerged in the process such as the distortion and exfoliation in the SU-8 spin coating on a non-circular substrate, the air bubbles in photoresist layer and the adhesion of plating layer with the substrate were eliminated or improved. Finally, the microinjection mold with a channel width of 80 μm and a height of 35 μm was fabricated. The results show that by using the ultrasonic stress relief technology, the SU-8 mould defects caused by large internal stress in SU-8 layer are overcome, the capacity of manufacturing microinjection mold via UV-LIGA technology is enhanced, and the success rate of microinjection mold fabrication is greatly improved.

The position precision and dynamic stability of optical elements for an off-axis Three-mirror Anastigmatic (TMA) space remote sensor are dependent on its main support structure, therefore, this paper researches the design scheme of the main support structure for the TMA space remote sensor. After analysis of the common forms of main support structure, a composing of truss support structure was determined, and its materials and connection technique were discussed. The analytical solution for the first order natural frequency of the structure with three struts was derived and the best one was selected. On the basis of a original design, the sensitivity analysis and parameter optimization were used to obtain a main support structure with 24 struts, then it was verified by the static reversal and a dynamic test. Test results show that the front rotation angle change of the main support structure is less than 10″and the first order natural frequency is 55 Hz, when it flips from a vertical state to a horizontal state. Furthermore, the maximum stress of the main support structure is 135 MPa during dynamic tests, which meets the design requirements of all indexes.

As the nonlinear hysteresis of a piezoelectric ceramic positioning stage affects its dynamic positioning precision badly, this paper proposes a model for the piezoelectric ceramic hysteresis based on a modified PI hysteresis model. In order to improve the dynamic positioning precision of the piezoelectric ceramic positioning stage based on traditional PID algorithm, the feed-forward control is constituted with the modified PI inverse hysteresis model and a traditional PID, and the experiments on slow speed and high speed positionings are performed. Experimental results show that the maximal error and mean error of the feed-forward controller are 40% and 20%-30% that of traditional PID algorithm for tracking single frequency trajectory.For multi-frequency trajectory tracking,the feed-forward controller has better dynamic control effect, and both maximal error and mean error are almost 33% that of traditional one.These data indicate the availability of the feed-forward control algorithm.

A novel composite compensation control scheme based on an improved Disturbance Observer (DOB) and fuzzy approximation was proposed to achieve the real-time stabilization and high-precision tracking control of an airborne opto-electronic platform. First, the kinematic coupling relationship between different frames was analyzed according to mechanical characteristics of the system. Then,an improved DOB structure based on the velocity signal was proposed to restrain the impact of carrier disturbance, and its basic principle and robust stability were analyzed. Considering the disturbance such as the friction torque that exists in mechanical systems generally, a composite control method with a fuzzy controller was proposed to improve the tracking performance. Finally, the global stability and the asymptotic convergence of the tracking error were proved on the basis of Lyapunov stability theory. Experiment results show that the stabilization accuracy can reach a higher level and the tracking error has been μrad magnitude, which demonstrates that the proposed scheme in this paper can restrain the impact of carrier disturbance and improve the tracking performance effectively.

A method of electrospinning was established to fabricate organic silicon gel submicrometer optical fibers. First, Tetraethyl orthosiliate(TEOS) and n-Octyltriethoxysilane (Octyl-triEOS) were hydrolyzed and polymerized under an acidity condition. Simultaneously, the fluorescence indicator of tris (4,7-diphenyl-1,10-phenanthroline) ruthenium(II) chloride ［Ru(dpp)3Cl2］ was added into the solution to form a viscous sol solution. Then, the electrospun fiber with a diameter of 900 nm was obtained by electrospinning the sol solution under a high voltage of 10 kV in the electrostatic field. A tapered multimode optical fiber was coupled with the electrospun optical waveguide fiber through an evanescent field and the sensing characteristic of the electrospun fiber was tested at the same time. Experimental results indicate that the electrospun (4,7-diphenyl-1,10-phenanthroline) ruthenium(II) chloride doped silicon gel fiber has a smooth surface and a uniform diameter. The light can be highly coupled into the electrospun fiber through the evanescent field and can excite the indicator molecules in the fiber to emit a 595 nm fluorescence. This fiber is sensitive to the oxygen gas in the range of 0~100 %(volume ratio),and the quenching degree of I0/I is linear to the concentration of oxygen in the range. The response time is below 100 ms.

A Continuous Multi-point Forming (CMPF) equipment was developed to manufacture 3D surface parts in high efficiency and flexibility, and the historical changing of forming loads and the influences of different technical parameters on forming loads were analyzed. First,the principle of CMPF was introduced,and theoretical formulas of forming loads were established based on certain hypothetical conditions. By taking a double curvature part as a research object,a Finite Element Analysis(FEA) model was established, and the historical changings of y direction load, z direction load and equivalent load were analyzed. Furthermore,the influences of press displacement, thickness, and curvature radius of a flexible roller on forming loads were analyzed. Finally,the equipment was designed, and experiments were carried out. Results indicate that the maximum value of y direction load is 6.693 kN, maximum of forming load is 6.716 kN, and the forming load is made up of y direction load nearly. Moreover,the maximum value of z direction load is 1.412 kN, which is considered as a driving force. Along with the increases of press displacement and thickness of sheet metal, the y direction load is increased. Along with the decrease of curvature radius of a flexible roller,the y direction load is increased. The changing situations of forming loads accord with practical situation, which offers a guidance for CMPF equipment. From experimental results, CMPF is a continuous, high efficiency and flexible forming method for 3-D surfaces.

To realize the non-contact inspection and treatment of human gastro-intestines, a control system with bi-directional Radio Frequency(RF) was designed for a micro robot to heat the cancer of colon. First, the modules of control system and its hardware configuration were established.Then, the software program was completed and the movement control was efficiently implemented according to the timing sequence presented in this paper. Moreover,a driving mechanism was used to simulate the inchworm movement and the generalized predictive control algorithm of PID was introduced into the heat therapy. The control of a micro turning table could provide 0~360° positioning of cameras to show the video of colon tube in rotation. Finally, the system performance was tested as well. Experimental results indicate that the average RF communication time is 0.6 s and the step of the robot movement is continuum and stable. The angle of camera view can be changed from 0° to 360°, and the heat treatment can be stabled in 70 s. The experimental results demonstrate the effectiveness and reasonability of this control system and it satisfies the system requirements of real time inspection and thermal therapy for colons.

In order to improve the production efficiency,surface quality and reduce the grinding margins of Polycrystalline Diamond (PCD) compact cutting tools, a series of experiments of Low-speed Wire Cutting Electric Discharge Machining (LS-WEDM) for a PCD compact were taken. The PCD compact was cut by the WEDM for five times,then the surface roughness of cutting section, the machining quality of cobalt-rich interface layer and the edge of PCD layer were measured and discussed. The results indicate that the PCD compact shows better surface quality in its cutting section after multiple processing by LS-WEDM, and the machining quality is affected by diamond particle sizes greatly. For two kinds of PCD compacts, CTH025 and CTB010, the surface roughness are Ra=0.85 μm and Ra=0.57 μm,and the depths of the grooves of cobalt-rich interface layer are 16.3 μm and 5.7 μm, respectively. The size of the defect in the edge of PCD compact is matched with that of the diamond particle. The minimum grinding margins of PCD cutting tools can be controlled in 4-15 μm after LS-WEDM.

An aluminum alloy workpiece was etched by a compound processing of laser and electrochemistry in an experimental system established, and the influence of mechanic effect caused by a laser shock and an electrochemical effect on the process quality was investigated .The results show that when a high-energy pulsed laser irradiates on the metal workpiece in the solution, it will take the elastic deformation by the jet impact and the shock wave force caused by the laser. Under the stress of the laser shock wave,the potential electrode is changed, current density is increased, and the electrochemical corrosion of metallic materials is accelerated. Moreover, the mechanics effect of laser shock could remove the passivation layer on the workpiece surface irradiated by the laser, and the material at this zone is eroded and removed. However, the other zone material could not be corroded with the protection of the passivation layer, which enhances the locally etching ability of electrochemical machining significantly. Finally, the compound machining of the laser shock and electrochemical corrosion was used to achieve a micro-etching with a high aspect ratio and a line width of 140 μm in the NaNO3 solution with a concentration of 0.5 mol/L, which obtains good processing quality and satisfactory shaping precision.

To achieve the ductile lapping of a single crystal sapphire, micro/nano mechanic characteristics of the sapphire (0001) plane were measured by nanoindentation and nanoscratch methods. The indentation model of single cone abrasive grain was proposed and then critical force conditions were deduced during ductile lapping process. Experimental studies were conducted for the single crystal sapphire based on the diamond abrasive grain charging into a synthetic tin plate, and characteristics of ductile lapped surface were measured by a NT9800 white light interferometer, a Scan Emission Microscopy(SEM) and a Transmission Electron Microscopy(TEM). Experimental results show that nanoindentation and nanoscratch methods can provide processing parameters for the ductile lapping of single crystal sapphires, and its critical depth of pile-up is around 100 nm for sapphire nanoindentation. The ductile lapping of the single crystal sapphire can be implemented by charging into diamond abrasive grains and selecting proper loads and the optimal load for ductile lapping is 21 kPa. After ductile lapping, the surface scratch depth of single crystal sapphire shows a smaller dispersion and the dislocation and slip are formed on the lapped surface.

A method based on optimization functions was proposed to improve the attitude measurement precision of a plane . First, the same feature that the plane was projected at two measuring stations was acquired by the template matching method. The coordinate value of the same feature for the plane was obtained by intersecting at a launching coordinate, and the initial value of the plane attitude was estimated in term of the characteristic triangle of the plane in space. Then, the object coordinate of the plane was established, the image coordinate according to the feature of plane was calculated by using a colinear equation and the least difference between the re-project result and an actual image was regarded as the optimizing goal function. Finally, the iterative method was used to improve the precision of the attitude parameter of a target. Experimental results show that the measuring error of the attitude angle is less than 0.1 ° when the axis of the plane imaging is more than 500 pixels. As compared with the middle axis method and the direction vector method of angle bisector lines, proposed method has a correct mathematical model, reasonable algorithm and can improve effectively the measurement precision of the plane.

The adaptive control scheme of a quadrotor Micro Aerial Vehicle (MAV) by using Interval Type-II Fuzzy Neural Network (IT~~IIFNN) was proposed to improve the control accuracy that was declined by the uncertainty, external disturbances,etc. Based on the quadrotor MAV dynamic modeling, an adaptive controller composing of two parts was designed by using the IT~~IIFNN,in which the IT~~IIFNN was developed to approximate the uncertainty function and a robust compensator was proposed to confront the approximate errors of IT~~IIFNN and external disturbances in real-time. Moreover,the Lyapunor stability theory was taken to prove the stability of the closed-loop control system in the quadrotor MAV.Finally, the superiority of the adaptive controller was verified by a prototype of the quadrotor MAV,which is shown that the tracking error approximated is 10-2 under the interference conditions of wind speed of 1.5 m/s.Experiments demonstrate that proposed control scheme can offer perfect tracking accuracy, stability and robustness.

An efficient iterative pose estimation method was proposed to improve the accuracy and robustness of monocular vision measurement systems. By using homogeneous coordinates,it avoids the limit of existing algorithms on the selection of reference points for a paraperspective camera model. First, a method to compute the pose of an object under the paraperspective camera model was researched using homogeneous coordinates, and its geometric meaning was expatiated in detail. Then, this method was used to estimate the pose of an object in an iterative way under the full perspective camera model with high precision. Simulation results indicate that the proposed method improves the accuracy, speed and robustness of the paraperspective camera model based iterative pose estimation. Real measurement results show that the translation measurement precision and rotation measurement precision of the proposed method are better than 0.1 mm and 0.1°,respectively,and it can satisfy the requirements of various vision detection systems.Furthermore, the accuracy of the proposed method can be improved further by using the land mark point method and sub-pixel feature locating tech-niques.夏军营(1983-)，男，河南驻马店人，博士研究生，2004年、2006年于国防科学技术大学分别获得学士、硕士学位，主要从事视觉测量方面的研究。

A thousand-unit scalable wave-front detector based image restoration technology was proposed. The wave-front detection and image restoration were combined to eliminate the effect of atmospheric disturbance and system aberration on the image resolution and to meet the requirements of large ground-based telescopes for high-resolution imaging. First, the wave-front detection was used to obtain the wave-front aberration and then degenerated images were restored based on obtained aberration amounts. The core component, a wave-front processor, used a wave-front host Printed Circuit Board(PCB) combined with a wave-front sub PCB to implement wave-front processing for optical systems with different sizes and it could reach thousand magnitude unit outputs. An image restoration experiment for a laser source was performed in an experimental laboratory, which shows the laser energy concentration has increased by 50%. And the binary source of 0.6 arc-second experiments was carried out in a telescope, and its Full Width at Half Maximum (FWHM) has decreased by 80%. By using a large-scale Field Programming Gate Array(FPGA) as the core processing device, the system achieves the high resolution image restoration for a degenerated image from the ground-based telescope.

A method to extract a light stripe central line with sub-pixel precision in real time was presented. First, the light stripe was segmented by light stripe contour tracking and the contour polygon representation. Then, based on the counter-clockwise sorting characteristics of contour and the orthogonal relationship between cross section direction and corresponding segmented light stripe direction, the scanning line direction of segmented light stripe was calculated by using a serial representation method. Finally, an adaptive barycenter method that can change adaptively the cross section width was designed on the scanning line to calculate the center sub-pixel coordinates. For the 640 pixel×480 pixel light stripe bitmaps captured by a 3D laser foot scanner with 6 CCDs at typical positions, the extraction time for central line is less than 3.1 ms in sub-pixel precision. The algorithm proposed here can satisfy the requirements of multi-sensor line laser scanning vision measurement system for real time, high precision as well as high speeds.

Several kinds of factors affecting the radiometric calibration of a large area array color CCD aerial mapping camera were presented through analyzing the process of camera imaging. Then,a model for radiometric calibration was established. The experiments of radiometric calibration were performed by using integrating sphere and in the dark field, respectively, and outputs of radiometric response of the camera were measured and analyzed through combination of changing the input radiance, exposure time, CCD gain and other factors. The experimental results show that the characteristics of radiometric response for R, G and B color channels are different when the input radiance is changed, but it varies linearly in the working spectral range. When the least square method is used to fit the calibration data of radiation, the goodness of fit for first-order polynomial shows that the R2 is above 0.999. Furthermore,when input radiometric luminance is constant,the output of the camera varies linearly with the exposure time or CCD gain, which lays the theoretical basis for the deduction of radiation response characteristic curves under the other work conditions. The data from radiometric calibration is applied to the exposure parameters, and experimental results indicate that the lightness of image is appropriate and its average gray value satisfies the requirement for exposure．

To obtain high resolution human retinal images, an adaptive optical system for retinal imaging was established. In the system,the residual aberration of the imaging system was used as initial parameter estimation for image restoration to get high resolution images by semi-blind deconvolation.First,the adaptive optical system consisting of a Hartmann-Shark wavefront sensor and a micromachined membrane deformable mirror was use to correct the dynamic human eye wavefront aberrations and to access the residual aberration of system when retinal images were capturing. Then,the optical transfer function was used as initial parameter estimation of image deconvolution modal to perform the iterative semi-blind constraint deconvolution on the retinal image and to eliminate the influence of residual aberration on imaging quality and obtain the high resolution retinal images. The experiment results show that satisfactory retinal image can be gotten by the proposed method. The image quality has been improved nearly once, and the success rate of imaging has raised from 38% to 78%. Meanwhile, the correction time reduces by 6/7. It concludes that the calibration time is reduced effectively, the success rate of imaging is improved, and the the scope of application is expanded by proposed method.

The non-uniform distribution of light illumination makes images show some shadow areas and highlight areas in the process of collecting images, which will lose a lot of image details. To solve this problem, this paper proposes a partial differential image enhancement method based on histogram equalization. First, this method transforms the image into a gradient domain, and the new gradient function is obtained through the gradient field transformation. Then, by using the least square principle to reconstruct the image from the gradient field, the textures and details of the image are enhanced and the unknown details are shown all. According to the non-uniform properties of light illumination, anti-balanced transformation is proposed by combining with the histogram equalization to adjust the brightness and contrast of the images. Furthermore, it uses the finite difference method to discrete the image. By combining with the heat equation, the method has characteristics of simple calculation, fast operation, stronger flexibility and wider applications. Based on programming by Visiual C++, its image processing time is 35 ms for an image with the size of 512 pixel×512 pixel, which meets the requirements of real-time image enhancement systems.

To obtain high spatial resolution images, a imaging system with two CMOSs was designed to acquire the same scene images simultaneously.Among them, one CMOS sensor was used to acquire high frame rate, low spatial resolution image sequences, and the other one to acquire low frame rate, high spatial resolution images.Firstly,the global motion path was obtained by computing the high frame rate, low resolution image sequences acquired by the CMOS sensor using optical flow method. Then,under the constraint conditions of energy conservation and the energy to be proportional to integration interval,the initial motion blur kernel was estimated and motion blur kernel was optimized by using alternating iterative method based on Bayesian criterion. Finally, a sharp image was restored quickly and effectively from the low frame rate, high spatial resolution blur image acquired by the CMOS sensor using TV-L1 algorithm. Simulation and experiment results indicate that more than 38% simulation images show their error ratios to be less than 2, and has little affected by noises.Restored images have smaller ringing artifacts,and the space-invariance motion blurred photographs can be deblurred effectively.

As the adjacent bands of a hyperspectral image are highly correlated, it is not optimum to classify the hyperspectral image in the high dimensional space. To solve the problem, a novel hyperspectral image classifier based on Steepest Ascent and Relevance Vector Machine (SA-RVM) was proposed in this paper. The SA was used to search an optimum feature space and to eliminate redundant features of the image firstly. Then, RVM was trained in the optimized feature subspace and used to classify the test samples. Experiments were performed for four sets of data,it is shown that the accuracies of RVM have raised more than 2.5% in the feature subspace selected by SA, which is close to those of Support Vector Machines(SVMs). For the two data sets with fewer training samples,the accuracies of RVM increase by 5.63% and by 6.2% in the subspace. In addition, benefiting from the sparse solution,the SA-RVM requires very short time in predicting the class labels of unknown samples. It concludes that the SA-RVM has higher precision and efficiency in the prediction, and it issuitable for processing the large-scale hyperspectral images.

As path loss prediction for a near-ground wireless channel is an essential issue in most research activities of wireless sensor networks, this paper measures and analyzes the transmission characteristics of a 2.4 GHz wireless signal in a grassland environment and performs a linear regression for collected data based on the least square method. The results indicate that the log-distance based model is still suitable for the path loss modeling in the near-ground wireless channel, and the fitting accuracy of the two-slope model is better than that of the one-slope model. Furthermore, the breakpoint position is determined by the antenna height, and the path loss indexes are less than 2 or between 3 and 4 before or after the breakpoint. With comparison of the predicted path loss and the measured path loss in a far-field, it proves that the prediction accuracy of the two-slope model is superior to that of the one-slope model. Finally, the performance of the generic models and the proposed two-slope model is compared, which suggests that the generic models are not suitable for research activities in the near-ground environment and the proposed model has a more practical value for research on wireless sensor networks.