In order to improve the ability for capturing and tracking the space flying object and to decrease the fluctuation of the real-time guide data with the intersection model switching,an intersection approach for multiple stations based on the least square estimation is put forward in the measurement network composed of two or more TV theodolites.According to the real-time geometrical relationship between the measured object and the TV theodolites,the proposed method makes use of TV theodolites position and their real-time angle information of the measured object to construct two perpendicular planes for every theodolite,which are intersectant at the measure direction line.The coordinates of the measured object is estimated by the least square principle.During the period of two theodolites taking part in real-time intersection calculation,the nonplanar intersection method is the special case of this method,so this intersection method is very effective.The simulation results show that compared with the nonplanar intersection method,the real-time guide data produced by this method is smooth and unfluctuating,the position error of the two methods is less 0.18 m,and the method can meet the needs for the multi-station TV theodolites intersection.

For the purpose of implementing carrier-based aircraft landing on the deck of carrier automatically under all-weather,a kind of infrared cooperative target group was proposed.Firstly,landing scheme based on cooperative target was arranged and a kind of cooperative target group was developed.According to the atmospheric window,transmission property of infrared irradiation in atmosphere and the impact of the most appropriate temperature corresponding to the optional infrared band on the controlling of the cooperative target's temperature,8～14 μm was identified as the radiation band of the cooperative target.The dimension of cooperative target was decided by detection distance of infrared imaging system and requirement.The result shows that landing points forecasting position error of three dimensions are less than 0.5 m,1.0 m,and 1.2 m,respectively.Opt-electronic landing guidance algorithm has the characteristics of high precision by using cooperative target group.

In view of the two disadvantages of current short-range missile warning equipment—one is that the scanning system can't get high quality image and the other is that the cost of staring system is too much,a scanning method about a kind of staring IRCCD detector was studied.The non-constant step-stare control method was applied to a small compound missile warning system.We compared conventional control means and step-stare control means,and focused on solving the application problem of the control technology in the IR scanning system.This project was used for a waning system.The distinguishing rate is 0.01° when the azimuth error is limited at 0.4mrad.The control method can be also applied to other warning detection equipments.

The problem of detecting small target in infrared video sequence was addressed.Instead of being regarded as prominence or high frequency part,the inherent isolated attribute of small target was taken into account and redefined as small isolated regions in image.Based on this idea,a new measurement called Small Target Isolation Degree(STID) was developed for evaluating the likelihood of a pixel as the center of a small target.By calculating STID in every pixel in image,the background clutter was suppressed.Then a novel adaptive threshold was proposed to extract the precise location of small target.Incorporated with the two methods,a single frame small target detection algorithm was built.Its high performance was then proved in a serial of experiments.

In solar adaptive optical system,absolute difference algorithm and correlation coefficients algorithm are widely used in Shack-Hartmann wave-front detection of the extended object.A Shack-Hartmann wave-front processor based on absolute difference algorithm is described in this paper.Real time wave-front processing of a system with 6×6 sub-aperture array,which is 32 pixels×32 pixels in size each and for which the reference image is 16 pixels ×16 pixels,is accomplished in a single Field Programmable Gate Array(FPGA) chip with absolute difference algorithm.The peak operation speed is over 23 000 000 000/s and calculation latency is 120 μs.Using this processor,frame rate of the Charge Coupled Device(CCD) can be up to 1 000 Hz.The design is realized in a single FPGA and it is low-cost,compact and easy to be modified.

Based on the geometric relation among orbital target,observation satellite,the earth,the sun,and the moon,an optical visibility determination model between observation satellite and orbital target is derived by taking into account of the earth background,the earth light,the sunlight,the earth shadow and the moonlight,and then a visibility prediction method for space-based optic observation is also given.The simulation result shows that the earth background and the earth light are the primary factors which influence space-based optic observation.On the aspect of observation time,it has limited efficiency for low earth orbit target space-based observation,and good efficiency for high earth orbit observation.

Aimed at the every key elements connected with the success of astronomical calibration,we performed the experiment research based on theory analysis to achieve the regular inspection and calibration of ship-borne optical measurement equipment.Firstly,we introduced the hardware principle of astronomical calibration technologies,software composition,and working procedure.Then,we focused on analyzing error separated arithmetic,and methods for solving error equation was brought forward based on singular value decomposition for the first time.Finally,a group of typical datum worked out from the experiments was shown.The calibration experiment results demonstrate that the calibration method can successfully separate 8 items of single error,and the repeatability of key items about many times of calibration is less than or equal to ±1″,which doubles the accuracy of actual angle measurement and is superior to the accuracy of normal calibration equipment.

Focusing on simulation of Spatial Heterodyne Spectroscopy(SHS),the emulational calculation of different characteristic light source is described.Based on the basic concept,configuration of system and the relationship between systemic parameter and interference,the conceptive simulations of monochromatic source and continuum spectrum source are discussed separately.The impact of different bandwidth and sampling bandwidth on the emulational results is also analyzed.Moreover,Contrast and analyze the experimental data obtained by Brassboard SHS and simulating result.These experimental results demonstrate that emulational data is coincident with experimental data,which offers us beneficial guide and necessary basis for designing and rebuilding SHS system.

Measurement accuracy of star sensor is restricted by ambient temperature.This paper describes an effective approach to analyze thermal effects in optical systems of star sensor.The star position errors,which are caused by thermo-optic effects,are calculated in an optical system of middle-precision star sensor.The results show that the star position errors are not remarkable near field angle 0°.And material matching is an effective measure in reducing the star position errors caused by changes of ambient temperature.After thermal compensation design,the maximal error is only 0.02″ when the ambient temperature changes from-20℃ to 60℃,which is about 1/7 of the original optical system's maximal error.

The Diamond-like Carbon(DLC) films were prepared by Pulse Vacuum Arc Deposition(PVAD) technique,the Laser-induced Damage Threshold(LIDT) of these films was investigated,and the preparation process was optimized.The results indicate that with the increase of the film thickness,the LIDT decreases firstly and then become stable above 100 nm.The study and analysis to orthogonal experiment results reveal that compared with other experiment parameters such as substrate temperature,cleaning time and pulse frequency in the experiment range,the main voltage is the most important factor to affect the laser-induced damage resistance of DLC films.An optimization process was obtained in the study at the cleaning time of 20 min,substrate temperature of 150 ℃,pulse frequency of 5 Hz and main voltage of 150 V.The LIDT of the DLC films can be improved by annealing in some degree.

According to the characteristic of Differential Optical Absorption Spectroscopy(DOAS) system,the Adaptive Neural Fuzzy Inference System(ANFIS) was proposed to evaluate the slowly varying components by using the nonlinear approaching theory of ANFIS in this paper.The network parameter can be flexibly regulated according to the measuring spectra.The testing results show that the ANFIS is more flexible than polynomial fitting and the fitting accuracy can be easily solved.The reliability and accuracy of DOAS evaluation have been improved for the application of the ANFIS.

The Differential Optical Absorption Spectroscopy(DOAS) technique uses the narrow molecular absorption bands and absorption strength to retrieve concentrations of the gases based on least square methods.We propose an algorithm for gas concentration recovery for the real world case where the DOAS system model is subject to uncertainties.The method is based on the formulation of concentration reconstruction as a regularization problem,where the dynamic change of gas concentration is treated as a state equation and the measurements of DOAS is used to form the measurement equation.The estimation is achieved with the aid of an adaptive Kalman filter framework.It is derived and extended from the Kalman filtering principles and is particularly powerful for real-world situations where the levels of the disturbances are unknown.The performance of our work is evaluated by using the experimental data.

To utilize the advantages of polarimetric imaging in target detection and recognition,the mechanism of depolarization in polarimetric detection and the influences of surface scattering and volume scattering on polarization degree are anayzed.Illuminated with 808 nm diode laser,the back scattered light of target is accepted by Cassegrain telescope.Two images are imaged on CCDs through Polarization beam splitter.Two orthogonal polarization state images can be obtained through once measurement,and then intensity and polarization degree can be calculated.Images coded with intensity and polarization degree are acquired.Experimental results show that intensity image can not distinguish targets when having similar albedo,whereas the polarization degree image can distinguish targets.Using HSI pseudocolor coding method to fuse the intensity and polarization images,the aim of target detection and recognition are achieved.

Extracting changed areas from different images was an important problem in the field of remote sensing image change detection.To solve this problem,a method based on maximal variance between-class criteria and C-means algorithm was proposed.Changed area extraction was converted into a typical problem of two-category classification and could be solved by employing threshold strategy.The C-means algorithm is used to classify an image into two classes and obtained its best threshold when the variance between-class is maximal.The experimental results show that the method can automatically determine the best image change detection threshold and extract the changed areas quickly and accurately.

The performance of passive Optical Current Transformer(OCT) with the linear birefringence is analyzed based on Jones matrix.The results show that passive OCT still works when the output light is circularly polarized,and the output voltage of passive OCT is affected by linear birefringence even though the output light is linearly polarized.The performance of passive OCT is analyzed when the Polarization angle and phase shift of the incident light and the different initial birefringence are introduced.The results show that when the initial birefringence exists,the relative deviation of system output has minimum.The corresponding incident light phase shift to minimum monotonically increases with the initial birefringence.A method of reducing the influence of passive OCT initial linear birefringence is proposed based on increasing the phase shift of incident light.The numerical and experimental results prove its feasibility.Developed practical optical current transformer has reference significance to improve the performance of passive OCT.

The theoretical acoustic sensing characteristics of a fiber laser are described based on Michelson interferometric demodulation technique.The fiber laser sensor is made by writing Bragg grating with π phase-shift into a section of Er-doped fiber.Acoustic pressure applied on the fiber laser induces laser wavelength shift.And a polarization-insensitive unbalanced Michelson fiber interferometer based on the 3×3 coupler is employed to transform the wavelength shift into phase shift of the interferometer.The signals are processed by Digital Signal Processing(DSP) to demodulate the phase shift.Based on the asymmetry characteristic of 3×3 coupler output signal,a novel algorithm for demodulating the phase shift is proposed.By employing two output signals whose amplitude can be adjusted in real time,it don't need uniform splitting ratio of coupler.The sensitivity experiment shows that the sensitivity is-166.5 dB(re 1 rad/μPa) when a 200 m path imbalanced readout interferometer is used,and the system output is linear to the pressure applied on the fiber laser sensor.

The work point should be controlled when open-loop Fiber Optic Gyroscope(FOG) is working,so a sin wave modulation signal with adjustable amplitude must be produced by digital method.First,mathematic description of FOG's modulation signal was analyzed,and then three methods for producing modulation signal with adjustable amplitude were discussed,which are direct wave table method,multiplicative amplitude adjusting method and two-level wave table method.Their advantage and disadvantage were analyzed.Experiment results of work point controlled by using different modulation signal producing method were given.In most cases,two-level wave table method is a good choice for producing modulation sin signal and can realize work point precise control in the condition of both limited storage resource and computing resource.

A novel method to detect early smoke is presented by processing the video for fire-alarming and gas leak-alarming.The whole detection procedure is mainly composed of four parts:1) off-line learning of the operation parameters of RGB color components;2) RGB color components operation combined with bit-masked color reduction to extract the smoke-like regions in real time;3) the background subtraction by background updating to remove those static disturbing regions,and then,the wavelet-analysis after the RGB color components operation to remove those dynamic disturbing regions;4) the multi-frames evidence analysis of long video sequence to check and label out the real smoke,i.e.all of the accumulated clues are combined to reach a final decision.Experimental results show that the proposed method can detect the early smoke correctly under a dynamic complex scene with the processing speed up to 16～20 fps for color image sequence of 320×240 resolutions,and the time needed for the decision-making is within 3～4 s.The novel method is robust to noises and is of high adaptability to monitor large and open spaces under a complex scene.

In the course of determining the aircraft model attitude in the spin time based on the stereo vision,the occlusion causes much trouble.A new way based on the credit coefficients is shown and it is mainly applicable to the problems with feature matching.In the occlusion problems,it is always the hardest to get the best optimal estimation of the occluded parts.The credit coefficients can weaken the effect of the features with poor definition.So the course for searching the occluded features is mainly based on the reliable features.Simulations and experiments on real images validate the reliability of our method and the results are good enough for the aircraft spin experiment.

To resolve the problem of mosaicing in portable cameras,a color difference space demosaicing algorithm based on all phase DCT/IDCT filters was presented.According to the correlation between the three color planes of color image and their sampling grids type,all phase diamond halfband IDCT filter was used to reconstruct the green image,and all phase extended DCT interpolation filter was used to reconstruct the R-G and B-G images.All phase IDCT interpolation filter was a pure 2-D filter which could convert between quincunx sampling grid and square sampling grid directly,while all phase extended DCT interpolation filter could restrain the Gibbs effect of FIR filters.Experimental results show that the Peak Signal-to-noise Ratio(PSNR) of reconstructed images outperform that of bilinear interpolation up to 6dB.And the proposed method can preserve more image edge details and has higher computational efficiency than other adaptive methods.

An approach for image segmentation is introduced based on region segmentation and edge detection.Edges are first detected in the original image and then cut equidistantly into some segmentations,and the end points of segmentations are gathered into a points set.A constrained Delaunay triangulation is yielded from the points set and then a graph is attained too.Graph cut,Minimum Spanning Tree(MST),is performed.Finally,the obtained regions are projected into the original image and region boundaries are refined to achieve pixel accuracy by Snake model.Experimental results show that the approach can achieve accurate edge localization,better spatial coherence and improved efficiency.

Endoscope images suffer from undesired luminance nonuniformity and loss of contrast because of variation of light resource intensity.The illumination-reflectance model and homomorphic filter are discussed for endoscope image enhancement.A novel homomorphic filter based on wavelet transform is proposed,which adopts fast 2D wavelet transform instead of Fourier transform.The wavelet analysis coefficients are processed by using a nonlinear mapping to attenuate the low spatial frequencies and amplify the high spatial frequencies.The experiments show that the new method can eliminate the non-uniform distribution of luminance of endoscope image effectively.Moreover,the method performs well in enhancing the local contrast of endoscope image,and its performance is better than that of the traditional homomorphic filter.

An improved C-V active contour model was presented,with which the image data force in the C-V model is significantly simplified according to the curve evolution theory.The level set function can be initialized to be zero according to the theory analysis.Distinguished from original C-V model,the level set function of the proposed model is Ordinary Differential Equation(ODE).The proposed model not only has the original C-V model's characteristics,such as the ability of detecting inner edge of the object with inner hole,etc,but also has many advantages as follows:anti-noise performance is good;re-initialization is unnecessary,and the efficiency of global segmentation is dramatically improved;the edges far away from the active contours would be seriously suppressed by setting the Dirac function to be one;the implementation does not require any limitation of time step.The method is validated with both synthetic and real images.Experimental results demonstrate the robustness and efficiency of the model.

Image edge extraction is the key step of video tracking and target identification.The performance and processing time directly influence the follow-up image processing performance and the overall visual system response time.A real-time infrared image edge extraction algorithm of eye movement mechanism is presented.The algorithm has features of strong edge recognition,but the computation is complex.Through the use of parallel processing structure and the pipeline technology,the algorithm is realized in a Field Programmable Gate Array(FPGA).The results show that the algorithm implements synchronization of real-time video streaming in a FPGA and have a good image edge extraction.Application of the image in real-time tracking system greatly increases the target tracking accuracy and stability.

For analyzing and avoiding the influences resulting from zero position error of control algorithm for controlling the exposure dose invariably on the Concave Spherical Substrate(CSS) via photolithography technology,the mathematical models and relation curves of zero position error versus scanning linear velocity and arc length are deduced and given respectively.After analyzing the relation curves,the influences resulting from zero position error are obtained as follows:vertical zero position error causes that line width of parts of lines are wider or narrower than ideal value and line width change monotonically,and the level zero position error causes that actual scanning arc length is longer or shorter than required value,which results in parts of edge of the CSS lacking of lines even grids.The opto-electronic encoders are adopted as feedback sensor of vertical and level axes,and the velocity and position dual closed-loop control method is taken,which realizes the zero positioning precisely and guarantees the exposure dose invariably,as well as scanning arc length accurately.Several eligible concave spherical meshes have been fabricated

Friction moment effect is one of main factors which affect the tracking accuracy of high-precision turntable system when the system works at a low velocity.The Stribeck friction model was put on the high-precision turntable system,and then a controller was designed,which could identify and compensate the friction moment online.According to the characters of the friction moment model,the controller was designed by feedforward+integral feedback strategy.And we analyzed the stability by using Lyapunov function and Lasalle common invariance principle,and the result we obtained indicated that the tracking system could achieve semi-global asymptotic tracking.Finally,simulation results prove that the controller can compensate the nonlinear friction moment online.

The jet characteristic of various nozzles with different structure was researched,and the design principle of nozzle was presented by the simulations of the jet characteristics of nozzles with various structures and geometrical parameters.The simulation results show that the fluid jet polishing system with the cone-shaped and columned nozzle can get a better jet characteristic and the geometrical parameters of nozzle affect the jet characteristics.When the shrinking angle is 13°and the ratio of length to diameter is 4,the issuing velocity of nozzle is well-proportioned and the turbulence intensity is low.This kind of nozzle is suitable to be applied in fluid jet polishing system.

Referring to the shortage and limitation of ruled grating which has ghost line and stray light because of period error and ruling irregularity,a method for making a Wavelength Division Multiplexer(WDM) is proposed based on Micro Electro-mechanical System(MEMS) blazed grating.The basic composition of WDM based on MEMS blazed grating is introduced according to the method.The craft for realizing MEMS blazed grating and the means for improving diffraction efficiency are also introduced.MEMS blazed grating is numerically simulated and analyzed in Laboratory Virtual Instrument Engineering Workbench(LabVIEW).And the diffraction intensity distribution of blazed grating is presented.At the same time,the rays in WDM are traced in ZEMAX.The results prove that MEMS blazed grating can be applied well in WDM.

To satisfy the requirement of encoder miniaturization,a fire-new encoding method and inimitable design structure of reading head are proposed to develop super miniature virtual absolute encoder.The new encoder adopted virtual absolute encoding method and single reading head.The new method combined distance-coded method with incremental encoding method and only used two rings in favor of miniaturizing encoder.This new encoder designed many absolute references in the whole disk,which quickly acquired current absolute position of code disk.Due to using single reading head,the technology of phase compensation was adopted to minish the size of structure.It has smaller size(Φ25 mm) and lighter weight,and the resolution can achieve 16 bits with processing error analysis and electric subdivision.