In order to study the variation of Transverse Chromatic Aberration (TCA) with external factors, such as the displacement of artificial pinhole from the visual axis and eccentricity,this paper constructs individual eye models to investigate the characteristic of TCA. The angle between visual axis and optical axis is calculated from the cornea data for eight eyes. According to the measurements of the corneal surfaces, optical axis lengths, wavefront aberrations and the angle obtained from calculation, the individual eye models involving the angle between visual axis and optical axis are constructed with ZEMAX software. The constructed eye models are then used to investigate the variations of TCA at the fovea with the displacement of artificial pinhole from the visual axis and to analyze the variation of TCA with the eccentricity over spectra from 420 nm to 700 nm. The result shows that the TCA increases with the displacement linearly at an average ratio of 5.7(′)/mm, but there are some differences among the eight eyes over the range from 5.46(′)/mm to 5.95(′)/mm. Across the field of view of 20° both at nasal and temporal,the TCA increases with the eccentricity at an average ratio of 0.36′/(°) for the eight eyes over the range from 0.32(′)/(°) to 0.44(′)/(°).Furthermore, the variation of TCA is slower at the end of long wavelength than that at the end of short wavelength. It is concluded that the influence of TCA should be considered in the designs of retina imaging systems and modern head-mounted display systems.

Stray light analysis on a three-mirror off-axis space telescope was presented in this article based on the measurement of Point Source Transmittance(PST). The entity model of the system was established to determine the first and second order scattering paths,and then different off-axis angles in ranges of 20° were calculated by the improved Monte Carlo ray trace method.Results indicate that the stray light mainly is composed of the first order scattering, which is related closely to the structure of the optical system. Furthermore, the PSTs are 3.56 and 4.02 at ±0.1° off-axis angles, 6.63×10-5 and 4.58×10－5 at ±20° off-axis angles. Experiments also show that the stray light can be reduced by modifying the baffles and changing the mirror’s scattering rate. Compared with that of the two-mirror off-axis telescope, the stray light of three-mirror system is graveness at the big off-axis angle, but it can meet the design requirements.

In a structured light system, coded patterns are the key of the corresponding point problem,for their reliabilities directly affect measuring accuracy and resolution. In the paper, an analysis of the maximum stripe deformation (due to depth changes on the surface) and measuring resolution limits are discussed and the principle of spatial periodicity used for coding is proposed. Results show that when spatial periodicity is used for coding, the resolution is greatly improved, or the number of patterns is greatly reduced for real-time structured light systems. According to these analyses, a novel coded pattern based on spatial periodicity for real-time structured light systems is presented as an alternative to the time-space coded pattern. The coded pattern allows range scanning of moving objects with easy implementation for decoding and high measurement resolution. Experiments demonstrate the validity of spatial periodicity used in coded pattern design.

A satellite borne Solar Backscatter Ultraviolet Spectrometer (SBUS) developed by our country has been put into operation in-orbit for the first time. This paper introduces the composition of the SBUS and gives its main working modes(atmospherical mode, solar modes and standard lamp mode). After analysis of run missions, a reasonable scheme of operation control in-orbit is presented according to the characteristics of various operation parameters and complicated mode transitions, and then the main points of software design is explained in detail. The remote sensing data of ultraviolet radiation measured in-orbit with the solar/atmospheric working modes are given, which shows that the wavelength reproducibility by the discrete solar mode is ±0.03 nm and the dynamic range of the system reaches 106 magnitude with the function of automatic gain control. Obtained results demonstrate that the proposed measurement modes have been implemented correctly and the scheme of operation control is completed. The SBUS are working in a best state with reasonable paremeters and optimum performance.

A tunable L-band ring Er3+-doped Fiber Laser (EDFL) with Mechanically-induced Long-period Fiber Grating (MLPFG) is reported,in which a 980 nm Laser Diode (LD) is used as pump source, and a 5×10-4 highly Er3+ -doped fiber is used as its gain medium. The optical fiber laser with a wavelength-tunable range of 42 nm(1 562.465-1 604.280 nm)is achieved by adjusting MLPFG fabricated period and altering the angle between fiber and grooved pressure plat. The 3 dB line-width and the 20dB line-width of the laser are less than 0. 04 nm and 0.08 nm respectively, while the Side Mode Suppression Ratio (SMSR) is more than 45 dB. Moreover, the power stability of the laser is better than 0.2 dBm for a prolonged observation. Experimental results show that the tunable EDFL has advantages of wide bandwidth, narrow line-width and stable performance.

In order to further correct the optical proximity effect in laser direct writing, the Iterative Error-Correction Method (IECM) proposed by Rajesh Menon is improved from two aspects. Firstly, one virtual acoustooptic modulator with high resolution is used to correct the direct writing pattern, and then the exposure data obtained are converted and rounded to that corresponding to the low resolution acoustooptic modulator to avoid the degradation during correction. Secondly, the exposure dose of writing spot that is the nearest to each error area barycenter is modulated by taking the minimum whole pattern error as a yardstick to achieve serial global optimization, which avoids a failure to correct the optical proximity effect when the number of positive errors is equal to that of negative errors. By taking the centerless block diagram as the test pattern, numerical simulation results indicate that the optical proximity effect of pattern with a low resolution acoustooptic modulator can be preferably corrected and the corrected pattern error value is decreased by 91.78% as compared with the initial error value. Furthermore, the optical proximity effect of pattern with a high resolution acoustooptic modulator can be further improved after being corrected by IECM, and the corrected pattern error value is further decreased by 1% based on 92.32% as compared with the initial error value.

With the aim to realize the miniaturization of absolute optical encoders with the characteristics of greater accuracy and more reliability, the new coding mode and structure of a reading head are investigated. Firstly, this paper describes the principle of eight-matrix coding used in absolute optical encoders. Different from traditional four-matrix coding,the eight-matrix coding only needs two grating tracks to realize a 10-bit binary.The first grating track realizes the 3-bit coding of Gray code; and the second one realizes the 7-bit coding of Gray code. Then it can achieve the 10-bit binary coding through the relation between Gray code and binary. The Moiré fringe obtained by using phase compensation of incremental slit is another new technique in this paper.The single reading-head is also good for reducing the number of LEDs, therefore the dimension of encoder is smaller. Finally, the method of signal pick-up is discussed as well. Experimental results indicate that the proposed eight-matrix encoder has a smaller size of Φ25 mm×16 mm and a lighter weight of 28 g, while his resolution is 16-bit and accuracy is 30″. It is concluded that the encoder can normally work in very hard conditions with more reliability for long time and can be widely applied to the field of modern military and aerospace.

The threshold selection method for the spot image of Shack-Hartmann Wavefront Sensor (SH WFS) has a large influence on the centroid detection accuracy. This paper analyzes the features of spot image of a SH WFS,and studies the effect of the threshold of spot image on the centroid detection accuracy,and then it presents an adaptive threshold selection method based on gray histogram principles. The experimental results show that the method can effectively and automatically select the threshold of each frame image when signal-to-noise ratio(SNR) of the spot image of SH WFS is more than 4, but it can not get the ideal segmented image when the SNR is less than 4. The SH WFS with the threshold selection method is used in the adaptive optics system to carry out a closed-loop correction experiment. After correction, the wavefront distortion of 10.41λ is corrected to 0.12λ in PV (λ=632.8 nm), and the system resolution reaches the diffraction limit. These results show the method can meet the practical requirements of SH WFSs.

In order to detect missile targets as early as possible, atmospheric absorption bands are used in the U.S Defense Support Program (DSP). To explore the band settings and their selection methods for the DSP,a new method to choose the optimal detection bands based on a background clutter model is proposed. The simplified parametric expression for a scene radiation is deduced, and then the spatial clutter model is established. The main factors impacting the clutter radiance are analyzed in absorption infrared wavebands,and the Synthetic Signal-to-noise Ratio(SSNR) is given in an analytic form. Finally, the optimal bands are chosen in all the possible scene clutter levels and target radiation characteristics, which provides the best system performance for detecting the same kinds of targets with the similar radiation characteristics under different clutter levels. The validity of the method is illustrated by examples. Calculation and analysis results indicate that the optimal detection bands are dependent on the scene clutter level and target radiation characteristics. For detection two different engine plumes, the optimal bands for DSP are 2.73-2.85 μm and 4.2-4.43 μm.

Aimming at the technological actuality of surface flow velocity measurement, a novel noncontact photoelectric measuring method for surface flow velocity is proposed. By deducing scattered light intensity formulae, it is concluded that scattered light intensities modulated by both up and down stream shallow sediments have cross-correlated characteristics based on sediment succession effect in shallow water.A formula for flow velocity measurement is derived and the effects of the measuring distance between up and down streams,flow velocity and sampling ratio on the measuring results are analyzed based on the cross-correlated characteristics.Then,an experimental system is set up, and the flow velocities are measured at 6 different flow rates in the cases of 6 sediment concentrations. The experimental results show that the measured flow velocity is coincident very well with the calibration flow velocity under six different sediment concentrations,which proves the feasibility of measurement theory. Experimental results also show that the absolute errors can be controlled to less than 8% and the correlative errors within 8% under the circumstances of suitable sediment in liquid.

A method for spectral data real-time acquisition with Firm shrinkage wavelet filtering was presented in this paper to improve the performance of biochemical analyzers. Firstly, two spectral lines were acquired and their average spectrum was taken to be processed. Then, the noise variance from two spectral lines was calculated to be used as the next threshold to replace the noise variance estimation based on traditional wavelet denoising. Furthermore, by adjusting the threshold on the line, the average spectrum was processed with Firm wavelet filtering and evaluation criteria were obtained. A relative spectral absorbance experiment was carried out based on the traditional 10 times average method and proposed Firm wavelet filtering method, and the experiment results indicate that the standard deviation of spectral signals obtained from the proposed method has decreased from 0.007 96 to 0.006 97,which means the acquired speed of the proposed method is higher than that of the traditional method. Moreover, the method has been used in a micro biochemical analyzer, tested results show that it optimizes the detection accuracy, reduces the detection time and the detection efficiency has improved by 4-5 times.

In order to fabricate 3D metal microstructures, a combined machining process based on UV-LIGA and micro Electrical Discharge Mechining(EDM) technology was proposed. Firstly,a quasi-3D metal microstructure was fabricated by UV-LIGA technology,and then a 3D metal microstructure was fabricated by micro-EDM. A 3D nickel mold with the local trapezium protrusion and taper groove was fabricated by this method and the sizes of the trapezium protrusion and taper groove were given in detail. Meanwhile, the influence of micro-EDM discharge parameters on the surface roughness was analyzed. The results show that the surface roughness is 0.08 μm, when the operating voltage and nominal capacitance are 65 V and 100 pF respectively. The research results demonstrate that the 3D metal microstructure can be fabricated by this method and the surface roughness will decrease by reducing the operating voltage and nominal capacitance.

A new method for measuring the aircraft’s 3D attitude based on intersection measurement，direction vector method of angle bisector lines, was presented. The coordinates of two suitable points in the edge character lines were acquired separately through the edge detection and character line for two digital images, then,the coordinates were transformed two times into a specific coordinate system. By combining the parameters of photoelectrical theodolite and tracked pitching and azimuth angles, the direction vectors of airfoil’s edge character lines in the specific reference coordinates were determined. Furthermore,by using the aircraft’s symmetry, the direction vectors of two angle bisector lines in the plane restricted by the two airfoil’s edge character lines were solved, and the vectors of axial line and its vertical line were reconstructed. Depending on operation above, the pitching, yawing and roll angels of the aircraft were obtained.An example for cobra maneuver flying tracked by two sets of photoelectrical theodolites in intersection indicates that the above-mentioned method is steady, valid and practicable. The error of angles can be limited within 1° if the image of airfoil’s edge character lines is clear. Obtained results show that the method not only provides a new way to deal with the optical measuring images, but also gives a reference for measuring poses of other moving objects.

Combined the Unscented Kalman Filtering (UKF)and the Unscented Particle Filtering (UPK), a Hybrid Unscented Particle Filtering(HUPF) was presented to improve the instability of the Micro Integrated Navigation System(MINS) for a vehicle when it was in a long-time moving, changes of dynamic performance and the case of GPS signal loss. Firstly, the Zero velocity Updates (ZUPT) was used to determine the dynamic characteristics of different vehicle systems. Then, according to the test results, the UKF or UPK algorithms were chosen to treat the performance of the vehicle under different dynamic characteristics. Finally, based on the experimental results, different ways were compared. It is indicate that the HUPF algorithm combined with ZUPT has reduced effectively the general error of the MINS of the vehicle under different dynamic characteristics, especially in the case of GPS signal loss. Obtained results show that the positioning accuracy and the error accumulation of the vehicle have improved by 55% and the processing speed has increased by nearly twice as compared with that of general ways.

With the aim to reduce the effect of the figure of a dome on the imaging quality of optical system for a missile and to decrease the flying drags of the missile, the figure of the dome was investigated and the outside and inside figures of the dome were optimized. Firstly, the Calculated Fluid Dynamics(CFD) simulation and a wind tunnel experiment were compared to obtain the accuracy simulated parameters, and the normalized relation function between the ellipticity and the drag of dome was derived by the CFD simulation. Then, the Optical Path Difference(OPD) was used to evaluate the normalized relation of the optical performance and ellipticity, and the optimized outside figure could be obtained after determining the weights of the aerodynamic and imaging quality. The obtained outside figure is an elliptic with R=37.5 and C=-0.75. On the basis of the outside figure and by taking the minimum deviation of the light through the dome as an optimizing goal, the optimized function was established by using ray tracing method, and an optimized inside figure of the dome was also obtained. The inside figure is a conic with R=33.31 and C=-0.78. An analysis shows the optimized dome has excellent aerodynamics and good optical performances.

In order to extract the central positions of a light stripe and to ensure the accuracy of surface measurement based on light-sectioning method, an approach of the light stripe central position extraction in sub-pixel is investigated. By taking the gray level of an image as a 2D function of coordinates, the directional derivative and the gradient of the image of light stripe are calculated.In the image, the pixel gradient represents the direction of the change in gray level, and it is assumed as the normal direction of the light stripe curvilinear. Pixels at the positions of light stripe can be extracted by setting a threshold for both absolute values of the gradient and the gray level. The sub-pixel central positions of the light stripe are equal to the mean value of the gray-level-weighed positions of the pixels in the stripe’s adjacent region along the gradient direction. With this method, the normal direction of the stripe can be simplified to be the gray level gradient, which avoids complicated steps to calculate a Hessian matrix to get normal directions. Experimental results indicate that the elapsed time of the method is only 10% that of the traditional way, and the error (3σ) of real measurement is 0.057 m in the range of 200 mm×200 mm. It can satisfy the requirements of accuracy and real-time measurement.

This paper describes the design, tests and preliminary results of detecting the X-ray pulsar vectors to acquire the attitude of a spacecraft by a multiple-plane sensor. Firstly, comparing to the star sensors or other traditional attitude measuring methods, a multiple-plane structure with the orthogonal vectors towards to the different directions is established and the multiple-plane sensor is set on the plane structure. Then, by measuring the photon power received by each plane and by considering the geometry relationship of the planes, the sensor is used to compute the vectors and spatial angle from a certain pulsar and to obtain the attitude of a spacecraft. Measured results show that the attitude of the spacecraft can be as much as 0.1° if vectors from several different pulsars are recognized and specified. The simulation is established and the result proves that the method is feasible in future spacecraft missions.

The tool can be worn seriously when Al/SiCp composite is machined due to the SiC particles with a higher hardness. Based on the new nano-cemented carbide characterized by his high hardness,toughness,flexural strength and higher wear resisting property,a WC-7Co nano cemented carbide tool was prepared, and the cutting experiment was carried out.The wear mechanism of nano cemented carbide tool and chip removal mechanism were studied. Results show that the wear mechanism of nano cemented carbide tool is an abrasion wear due to the micro-cutting of SiC particles and the micro-chipping and micro-breakage of cutting edge caused by interrupted impact of the SiC particles under the high frequency. The essence of cutting Al/SiCp is interrupted cutting,and the Al/SiCp chip removal is the breakout of chip.Experiment indicates that the flank wear value of nano cemented carbide tool is about 30%-50% less than that of common cemented carbide,which means nano cemented carbide is more suitable for machining Al/SiCp composite than common cemented carbides.

In order to overcome the significant impact of hysteresis on positioning precision when Giant Magnetostrictive Material Actuators(GMAs)are applied to fine positioning, an adaptive PID control method is studied and a multimode PID control method based on Generalized Predictive Control (GPC)principle is established under the condition of the hysteresis compensation. Firstly, the relation between the parameters of PID controller and the parameters to be estimated is derived from GPC principle, and GPC-PID control is realized. Then, for the quibble problem at the beginning stage, the control mode switch condition of multimode PID control is put forward. Finally, a multimode PID controller for the actuator is setup to switch the control modes from the regular PID to the GPC-PID in real time based on PID parameter variability. Experimental results indicate that although the GPC-multimode PID controller extends the consuming time of each execution step slightly by 7 ms, the RMS of tracking error is reduced by 0.066 μm as compared with a Generalized Minimum Variability-Fuzzy PID (GMV-Fuzzy PID) controller, and at the same time, it improves the stability of the beginning stage. These results show that the multimode PID controller can overcome the impact of system disturbance, smooth the beginning status and can be suitable for the precision positioning with high accuracy and real time.

A new microfabrication technology based on PMMA(polymethylmethacrylate)microneedles was presented by using the PCT technology with X-ray and lithographic mask. By moving lithographic mask, three-dimensional PMMA microstructures could be made,and the fabricated shape was directly dependent on the absorber pattern on X-ray mask. From the experiment, the final microstructure was not the same as the absorber pattern on X-ray mask. Experiments show that if there is no compensation for mask design, the shapes of sloped side-wall on the exposed structures will be deformed, which affects the performance of microneedles. The main reason of the deformed shapes of sloped side-wall was analyzed in this paper,and results show that it comes from nonlinear relationship between developing time and exposure energy. Based on the PCT method of fabricating PMMA microneedles, the length of the microneedle is 100-750 μm, and the diameter is 30-150 μm, which means that the minimum of the microneedle’s diameter is 100 nm. Based on the compensation of the mask pattern, the strength of microneedle with the channels has been improved by changing the mask pattern from a hollow double right-triangular pattern to a hollow double semi-elliptic pattern.

In order to control the cylindricity errors of small holes machined by Electrical Discharge Machining (EDM), the relations between the electrical parameters and the cylindricity of small holes was studied by technologic experiments. In the course of studying, a multi-cross section measuring method in a Cartesian coordinate was proposed. The Wire Electrical Discharge Machining(WEDM) was used to cut the processed workpieces in axial and radial directions to obtain the generatrix datum and circle datum of machined small holes, then the characteristic points of two data were measured by an universal tool microscope, and the cylindricity quantitative evaluation of the small holes was realized. Based on the method above, an orthogonal experiment was carried out, and the effect of some main electrical parameters (voltage, current, pulse duration, pulse interval and tool-lifting) on the cylindricity of small holes was studied. The results indicate that the pulse width affects cylindricity most distinctly, the voltage comes second, and the effect of rest parameters show a bit weaker.A sets of discharge parameters are achieved to purchase the least cylindricity error, which are V=110 V, I=0.6 A, ton=150 μs, toff=120 μs and tool-lifting time=1+2 s. In the experiment, a small hole with a average radius of 0.301 84 mm can offer the cylindricity of 0.021 mm,which means the presented method is effective to control the cylindricty of the small holes machined by EDM.

Cone-beam helical Computed Tomography(CT) can scan a long object, but the conventional scanning mode of X-ray CT imaging is limited by the width of planar detector when a large object is inspected. As an alternative, a scanning and reconstructing algorithm is presented based on the Field of View(FOV) for half-cover scanning to extend the diameter of the FOV for the cone-beam helical CT. In scanning, the gantry is translated horizontally, whose direction is vertical to the central ray and rotary axis. Then the conventional helical scanning is used to obtain the projection data needed. During reconstruction, the extended FDK algorithm can provide the same efficiency as the conventional algorithm without resetting the projection data. Experimental results indicate that the half-cover scanning can extend the FOV up to 1.86 times that of the conventional method, and the image quality reconstructed by improved algorithm with a small planar detector is almost the same as that of the conventional algorithm with a large planar detector. The reconstruction time has reduced by 376.66 s due to the decrease of projections. These results show that the half-cover scanning of helical cone-beam CT can extend the diameter of the FOV, and can offer a higher calculation efficiency and low reconstruction time.

In 3D edge detection for an industrial Computed Tomography(CT) image, the Facet model can be used to obtain the relatively precise image edge, but it will cost too much time. Otherwise, the wavelet transform method can be taken to detect 3D image edges in a shorter time, however, the edge information obtained depends on the threshold strongly . In combination with the characteristics of two methods above, a 3D edge detection method based on wavelet transform and Facet model is presented in this paper. Firstly, 3D wavelet transform is used for an industrial CT 3D image , and initial candidates of edge could be obtained by setting a smaller threshold. Through this step, the edge is located roughly. Then, 3D Facet model is used to decide if the candidate is the true edge, and the points in rough edge are fitted one by one to realize the precise location of the image edge. The improved algorithm reduce the number of edge pixel candidates by the preprocessing of 3D wavelet transform, which improves the detection speed of the original 3D Facet model. Experimental results show that the proposed method not only can obtain the same image edge as that by original 3D Facet model but also can improve the image detection speed greatly. Obtained results indicate that the image detection speed has increased by 3.51-7.39 times as compared with that of original 3D Facet model, and the simpler the image edge is, the higher the accelerate ratio is. These results show that the 3D edge detection method based on wavelet transform and Facet model can meet the requirements of the 3D image in an actual industrial CT for high accuracy and high speed.

A crack detection method of Industrial Computed Tomography(ICT) noisy images based on fast Beamlet transform is presented in this paper. After analyzing the composition and relation of Beamlet in a mono-scale, a fast Beamlet transform is proposed. On the basis of the Beamlet transform, a control variable about the relativity is introduced. Then, combining the tree-structure of Beamlet’s mutiscale and a “top to bottom” inter-scale inhibition to optimize the object function, the crack is detected. Finally, considering the near pixels of detection result of the crack, the edge of crack domain is extracted. A numerical experiment on the images including an original ICT noisy image, a Gaussian white noise image with a variance of 0.1 and a salt & pepper noise image with a superposition density of 0.1 is carried out. Compared with the methods of Laplace, Canny or wavelet, the proposed method can detect the crack of ICT noisy image more effectively. Because the Beamlet transform uses lines to analyze the image data, the proposed method has a robustness to noises.

Triangle identification algorithm for star images has been widely used in attitude detection systems that take star sensor as navigation parts at present.However,its low characteristic dimensions lead to a higher redundancy and a lot of mistake identification. To improve the accuracy of triangle identification algorithm, a modified algorithm based on traditional one was proposed. By the proposed algorithm,the guide star pairs that meet interstellar angle threshold were recorded in a star matrix and the times appeared of the guide starts were counted simultaneously. According to the truth that each star must be attached to two sides of the triangle, the special star pairs that appear less than twice should be eliminated, then the guide stars for contructing triangles were searched from stars left in the star matrix. The algorithm takes the interstellar angle and relative magnitude of stars as the identification feature and also take the hashing search method into the star identification, therefore , it reduces the comparison times and increases the identification speed. Furthermore, it also decreases the capacity of guide star database.

A region-margin line segment stereo matching algorithm is proposed to improve the posture recognition precision of a manipulator. The color characteristics of the region and the geometric characteristics of margin line segment included in the image are used to realize an integrate matching，among which the former contains the color and margin line segment，and the latter contains lengths，directions and angles.By making full use of the information of color included in the image ，especially the gradient direction information, this algorithm makes the line segment matching depend on the all information of the image area,not only on its geometric characteristics, such as lengths，directions and angles. Therefore，the criterion of the matching is more accurate than those of traditional algorithms.A posture recognition experiment for a manipulator is undertaken,obtained results validate that the algorithm could recognize the posture of manipulator in higher accuracy in a complicated background, and the relative error reaches 1.7%,which meets the requirements of the attitude recognition of manipulators for the precision.

In order to improve the visual field of photos, an optimized algorithm of automatic image mosaic is presented based on frequency and time domains. The frequency phase correlation is adopted to sort the unorder image sequence and to estimate the overlapping area, and an improved Harris operator is adopted to extract corners. Then, the bidirectional greatest correlative coefficient is used to obtain the initial feature point pairs, and the false feature point pairs are rejected by RANSAC. Finally, the image mean combined with the linear weight function is used to implement image mosaic. Experiment results show that this algorithm can offer a excellent sorting process and show good nomosaic results.The maching speed by proposed algorithm has improved by 40% as compared with those of traditional methods,which efficiently solves the difficulty in confirming corresponding points and achives the desired visual effect in mosaic images with notable illumination difference.

An improved dynamic programming algorithm was presented to detect space small targets from sequence images. Firstly, the development and current situation of the dynamic programming algorithm used in small target detections were introduced. As the defocus was used to process commonly a project to obtain the sub-pixel center of a target,the recursive equation of score function was improved.A multi-point accumulation algorithm with direction weight was presented to replace the single-point accumulation in an original algorithm, and the original multispeed plane calculation was also simplified. Finally, the key processes including state initialization, constant false alarm rate threshold and track data structure were discussed. The analysis indicates that the computational complexity has reduced about 50% as compared with that of original algorithm. Experimental results also indicate that targets can be detected in 5th frame with the multi-point algorithm, while a number of fake tracks are found until the 10th frame with the original algorithm. Furthermore,the score SNR of 20 tracks with maximum score in the 5th frame of multi-point algorithm is 0.87 higher than that of original algorithm. The multi-point accumulation algorithm with direction weight is able to reduce the false alarm caused by a speckle noise efficiently and to raise the detection ability of dynamic programming.

Architecture-based reliability models for the software reliability evaluation are tested and verified in this paper. To analyze the failure behavior and the relationship between software modules precisely, the architecture-based models are classified as a composite type and a hierarchy type according to the different evaluation methods. Then, how to determine the software architecture and to divide the software models is disccussed and the estimation methods and processes for software reliability and transferring probalilities between the softwares are given. A control software for a ground target is analyzed and verified,results indicate that the assessing value of composite and hierarchical models are 0.972 and 0.969,respectively,which is closed to the actual reliability of 0.938. The method can provide a references for the implementation of architecture-based software reliability evaluation.

It is necessary to obtain the depth information of a cement clinker for distributing the cooling air in a grate cooler reasonably. Considering the high-temperature in the grate cooler, a mesh-candidate-based online 3D measurement method was proposed to reconstruct the the cement clinker. A binocular stereo system composed of two cameras with the optical axes intersected was used to capture the images of the cement clinker and a wavelet multi-resolution algorithm was adopted to improve the efficiency of the measurement system. If the confidence coefficient value is lower, a few candidates should be rematched using a larger or smaller windows according to the histogram, by which the measurement error caused by ambiguity could be reduced. Experiments indicate that the root-mean-square error of the system is 4.29 mm, which meets the requirement of the measurement precision. The measurement method shows its advantages in simplified implementation, high accuracy and low costs,and it can be stably applied to the production line of the grate cooler.

A Automatic Gain Control(AGC) algorithm for Solar Blind Ultraviolet Enhanced ICCD(SBUV-ICCD) was proposed to detect the targets with different radiation luminances and to avoid damaging the device when it was exposed to high intensity radiation. The sum of N consecutive frame histograms were treated as an analysis object, and it was divided into background section, target section, brightness section and saturated section. The AGC algorithm was accomplished by using the proportion factor of brightness section to target section and the coefficient of saturated section to brightness section as controlling parameters. Finally, experiments were designed for testing the algorithm performance. Through controlling saturate ratio below 0.2 and bright ratio in the range of 0.4-0.8, experimental results indicate that the algorithm can adjust the gain of ICCD quickly and effectively in 1-3 s whenever the target's luminance is up unexpected rapidly or changed slowly, obtained results show that proposed algorithm can offer a good image resulution while the camera is protected.

In order to realize sub-aperture stitching measurement of a large-aperture optics, a work-piece localization method based on stereo vision is introduced,and a measurement system is set up. Firstly, the measuring principle of a sub-aperture stitching test is given, and its requirement for the work-piece localization is analyzed. Then, the stereo vision measurement system is introduced and its measuring model is established. After that, the marked points’ 3D coordinates which are rigidly connected with the optics under test are obtained and the transformation matrix is solved out by the quaternion method. Then, the full-aperture phase map is synthesized by mathematically minimizing the mismatch among the overlapping areas of all sub-apertures. Finally, a 150 mm flat and a 100 mm convex sphere are tested using this method. Experimental results indicate that the translation precision and rotation precision of the system are better than 0.1 mm and 0.01°, respectively,which can provide an effective initial value for the stitching algorithm.Furthermore,the measurement error of the system does not accumulate with the motion in its view field,and can satisfy the system’s precision requirements.