A portable nonmydriatic digital retina camera in 30° field of view was designed based on Gullstrand-Le Grand eye model.By considering the aberrations of both human eye and camera,this design realized the highly clear image with 2 mega pixels in all fields of view.And an eye model was used in this design to diminish the chromatic aberration caused by the human eyes themselves,so that the camera could be used in a visible band.Meanwhile,the illuminating system with an annular stop and coaxial illumination was specially designed to remove the stray light caused by a cornea.The tested result shows that the resolution of this system is higher than 120 lp/mm,and the distortion value is only 7.2%,the field curve value and the chromatic value are less than 0.86 mm and 3.5 μm,respectively.It comes to a conclution that this camera has strong focusing abilities,and adapts to different eyes from -10 m-1-1.

The radiometric calibration for UV-ICCD was investigated to build the relationship between the input irradiance and the digital output.The theory of radiometric calibration was derived and the radiometric calibration of the UV-ICCD was implemented based on spectral irradiance standard deuterium lamp with uncertainty of 5% calibrated by NIST in a laboratory.Under the conditions of fixed MCP gain and fixed integration time,the response of the UV-ICCD was tested and the relationship of the response and the MCP gain was studied.The preliminary analysis of calibration data shows that the response of UV-ICCD is linear and the MCP gain is directly proportion to the average grey value of image.Finally,the uncertainty influencing the calibration results was analyzed,the analysis shows the maximal uncertainty is 7.94%,which can meet the calibration requirement less than 10%.

A Φ100 mm deformable polishing lap consisting of 19 PZT actuators was designed and manufactured to polish a hyperboloid mirror with Φ=350 mm,k=-1.112 155,and R=840 mm.In order to analyze the influence of location error on surface error,the formulas were deduced to calculate the deformation and deformation errors of any points on the lap surface.The surface errors were calculated by finite element method,calculated results show that when the distance from deformable polishing lap to the hyperboloid mirror center L is 120 mm and location error δL is 0,1.0,-0.5,0.5,1.0 mm,the surface error RMS are 0.329,0.454,0.366,0.367,0.461 μm,which increases by 0,38%,10.9%,11.2%,40% respectively.These data given here suggest that the location errors should be controlled less than ±0.5 mm.

An ex-situ determining method was proposed for the optical constants of aluminum thin films precisely.Compared to the traditional in-situ method,the method had no need of the complex optical measurement apparatus attached to vacuum chamber and could do measurement simply and fastly.By using multi-wavelength fitting method,the ellipsometric and photometric spectra were fittd with a combination of optimal genetic algorism and a simplex algorism and the oxide layer formed on the aluminum in air was considered in the fitting process.With the combination of parallel genetic algorithm and least square method,the fitting speed and quality were obviously improved and the optical constants from the UV to visible range were obtained within an error no more than 2%.With these data known,an UV induced transmission filter with one cavity was designed.The tested transmission spectra are well consistent with that design in a peak transmittance error less than 1% at 265 nm.

An auto-tracking and guiding system for full solar disk image was introduced based on mass-center algorithm and large CCD-array.An embedded system with high speed processing ability was used to compute and process mass-center data and all CCD images were dealed with binarization process.Compared with the traditional system,the new system can greatly reduce the cost and complexity and can enhance the accuracy of the tracking system and reduce the dependency of weather condition.A high-precision closed-loop tracking apparatus was designed also in this system.According to the analysis of the experimental data,the new auto-tracking and guiding system achieves the precision of 1″ during one hour observation,which satisfies the requirements of the system.

In order to design the beam-shaping system of Laser Diode Arrays(LDA) and to evaluate the beam quality of LDA availably,effects of non-luminescence of Bar on beam properties of LDA were studied.Based on farfield bi-peak structure of an emitter,a far-field expression model of LDA was proposed.The simulation results agree well with the measured data and the overall error of this model is less than 5% in the more than 95% power region.On the base of this theoretical model and measured data,effects of non-luminescence of Bar on intensity distribution and other properties of LDA were analyzed.The results show that non-luminescences of single Bar or several Bars can not change the far-field bi-peak of LDA after beam transmission exceeds a certain distance,but it can effect on other properties such as peak intensity,the position of peak intensity and the intensity on center axis.The power of LDA is inversely proportion to the number of non-luminescent Bars and do not be effected by position variety of non-luminescent bars.

SiC:H surface-modified coatings were fabricated on Reaction Bonded Silicon Carbide(RB SiC) substrates.Silicon was evaporated by E-beam and methane was ionized as reactive gas by End-Hall ion source.X-ray Diffraction(XRD) results show the fabricated film is α phase,its hardness and elastic modulus are in the range of 9.781～13.087 GPa and 89.344～123.413 GPa,respectively.After surface-modification polishing and coating silver,the reflectance of modified RB SiC film is close to that of fine polished Zerodur glass coating with silver.Moreover,the fabricated coatings to the substrate show very good adhesion and no fall-off and cracks in thermal impact test from liquid nitrogen temperature to boiling water temperature for 5 cycles.

Based on phase Shifted Superstructure Fiber Bragg Grating(SSFBG),the structure and bipolar encoding/decoding principles of an Optical Code Division Multiple Access(OCDMA) system were researched and its correlation property was simulated by transmission matrix method.The effect of Bragg wavelength deviation due to the mismatch of the grating period between encoder and decoder on correlation property was analyzed theoretically.The results indicate that the correlation property is improved with the increasing of code length,but it is more sensitive to the deviation of Bragg wavelength between encoder and decoder when the code length is increased and the chip pulse width is fixed.To obtain desired correlation property,the deviation of Bragg wavelength between encoder and decoder must be less than 0.02 nm for the Gold sequences with length of 31 and chip duration of 6 ps at 1.555 μm.It is conclusion that a code can be used in different Bragg wavelengths to implement hybrid Wavelength Division Multiple(WDM) OCDMA system.

In order to improve the coupling effect of focused laser and waterjet fiber,a new coupling system consisting of an accurate coupling alignment system and a special coupling unit was designed,and a special coupling alignment method also was put forward.Before laser arrives at water-jet fiber,it passes through air,glasses and water,so a laser burning method was used to determine the distance between focus lens and nozzle hole based on calculating result.The max angle of laser incidence was researched for full reflection.After fluid simulation,the coupling unit was designed to make the fluid field in the chamber symmetrcally and form a high quality water-jet fiber with diameter of 0.12 mm.The attenuation of laser energy in water-jet fiber was investigated and the energy attenuation in fiber was reduced applying appropriate laser wavelength,pulse energy and filtered and deionized water.The experimental results indicate that a 100 mm long water-jet fiber is formed in this paper,and an 0.12 mm wide slot without nearly crack and heat affected is cut in 0.2 mm thick Si wafer at the speed of 2 mm/s,which shows proposed coupling technology can meet the requirements for water-jet guided laser micromachining perfectly.

After introducing to the inner structure and the driver timing of a interline area-CCD KAI-1010M briefly,a high speed and low cost CCD driver circuit used in digital space survey camera was designed in a new way.In order to resolve the technical difficulties,the pulse width modulation technology usually used in power supply and motor control was introduced into the CCD driver circuit and a ultra-high speed operational amplifier was adopted to reduce the rise time and fall time of high speed negative signal.Experimental results show that the CCD driver system can output two groups of CCD signals simultaneously at data output speed of 15 frame/s,which can meet the requirements of the camera system because of its good function and low cost.If the circuit is improved further,the CCD driver system can output two groups of CCD signals at data output speed of 27 frame/s.

In space environment,the micro-lens of interline transfer progressive scan Charge Coupled Device(CCD) will become fuzzy,for the effect of space ray in a long time results in bad quality image of the space camera.However,an interline transfer area CCD without the micro-lens has a lower fill factor and quantum efficiency,it will result in bad quality and lower ratio of Signal to Noise(SN) of image also.In order to apply this kind of CCD in space camera,a Time Delay Integral(TDI) pattern to overcome the conflict based on the character of the CCD was proposed,a system using Field Programming Gate Array(FPGA)to generate the driving signal was designed to realize the TDI pattern.At the same time,the communication and data transfer of FPGA with computer was completed by RS485 bus for setting the TDI stage,the integral time of TDI and the gain and offset parameters of video process chip.The experimental results indicate that the SN ratio is only 22.13 dB(the TDI stage is one) and the quality of the image is very bad when imaging without a micro-lens.After setting a reasonable TDI stage parameter,the SN ratio of the image is 33.62 dB(TDI stage is four),the quality of the image has a big improvement.Experiments show that the TDI pattern of interline transfer progressive scan CCD can improve the quality of image without micro-lens under space ray environment,Which resolves the conflict of micro-lens can improve the quantum efficiency of interline transfer progressive scan CCD,but micro-lens will become fuzzy for the effect of space ray in a long time.

In order to meet the requirements for high reliability of space camera controller,a self-testing method of Digital Signal Processor(DSP) space camera controller based on mixed language programming was proposed.With this method,the difficult problem of changing variable locations in RAM according to self-testing results was solved under mixed language programming by modification of runtime-support library.Also,in this way,the image motion compensation algorithm could be realized easily in DSP space camera controller.Experimental results show that the maintainable and transplantable abilities of the codes are improved greatly.Several problems on realization of self-testing arithmetic on DSP repertoire,Read Only Memory(ROM),Random Access Memory(RAM) and other peripherals were analyzed and solved,and an improved peak value test algorithm of RAM was put forward,so that the probability to detect RAM fault with improved algorithm is at least twice as large as that of original algorithm in the same running time.As it is difficult to test self-testing algorithm itself without harming hardware,a method to validate self-testing algorithm was presented.The results of validation experiments and practical application effects in engineering prove that the self-testing algorithm is practical and valid.

In order to reduce the influence of outside load on the surface precision of primary mirror in a space remote sensing camera,a design idea of adopting diverse flexible support structures to restrict motion freedom of the primary mirror independently was investigated.By adopting CAD engineering analytical software,the dynamic and static rigidities as well as thermal characteristics of a primary mirror system were analyzed.Based on this foundation,the structural parameters of flexible support in the primary mirror system were amended,so that the influence of gravity,assemblage stress and temperature diversification on the surface precision of the primary mirror was reduced under precondition of ensuring support rigidity,and endurance of impact and libration was ensured during the launch.Detecting results show that the surface precision of primary mirror reaches PV:λ/4 and RMS:λ/50,which can meet the special requirements for the primary mirror of space remote sensing camera.

For purposes of designing a mapping camera with high image quality for stereoscopic mapping task,the design,manufacturing and assembly for the lens of a mapping camera are investigated.the design of the lens is issued from material choice,structure decision,performance analysis,and so on.Then,the processing methods for thin shell supporting structure of the lens are discussed.Finally,the assembly process of single lens unit and primary lens is introduced as well.After comparison,ZTC4 is selected as the material for lens tube,and a single lens unit and a thin shell are adopted to a combination format.Analysis results with PATRAN/NASTRAN indicate that the lens shows a good mechanical intensity and stiffness and can satisfy the requirement of structure of single lens.The thin shell is gained through casting and turning precisely,and the co-axiality of column in orientation ladders is 0.01 mm.During the assembly of lens,it is very important to keep no deformation,to control a air space accurately and to position a high precision center for assembly.The tested result shows that the optical transfer function of single lens can reach to 0.43(77 lp/mm),which can meet the requirements of precision for a mapping camera.

In view of the features of an on-board embedded software,an on-board embedded software testing process model,butterfly model based on W model and X model,is proposed.This new model not only reflects the test for entire development activities,but also supports iteration and alteration.Especially,it is suitable for the characteristic that on-board embedded software development process needs multiple cycles.This on-board embedded software testing process model has many advantages in finding the design defects of software and hardware at early stage for satellite products,shortening the development cycle of on-board embedded software,reducing the costs of software and hardware and effectively improving the reliability of on-board embedded software,and it is suitable for the whole development process of on-board embedded software.

A new calibration method for Extreme Ultraviolet(EUV) monochromator was presented.By measuring the 30.38 nm and 58.43 nm lines of a standard He hollow cathode source and the reflectivity peak of a calibrated Mo/Si multilayer mirror with central wavelength of 13.90 nm,the wavelengths of Mcpherson247 EUV monochromator were calibrated.Then,the standard points on the track of monochromator was fitted to get parabolic curve by Origin.After analysing calibrated results,it shows that the measuring accuracy is 0.08 nm and the repeatability is ±0.04 nm,when the reflectivity peak of the multilayer mirror is measured using soft X-ray reflectometer with laser-produced plasma source at 12～60 nm.The main sources of measurement errors are the instability of the light source and the rotation error of the mechanism.

To satisfy the requirements of 13.9 nm Ni-like Ag and 19.6 nm Ne-like Ge X-ray laser researches,Mo/Si multilayer mirrors were prepared,and a small and simple structure reflectometer easy to operation was designed.A reflectivity-measuring system consisting of a reflectometer,a Mcpherson 247 monochromator and a Extreme Ultraviolet(EUV)radiation source from copper-target Laser Produced Plasmas(LPP) was established.The calibration precision of the monochromator was 0.08 nm and the intensity variance of the LLP source was lower than 5%.The experimental results show that the reflectivities of Mo/Si multilayer mirror at 13.9 and 19.6 nm are 41.9% and 22.6%,the related FWHMs are 0.56 and 1.70 nm,respectively.The measurement of surface roughness was carried out using WYKO,which are 0.52 nm for 13.9 and 0.55 nm for 19.6 nm.

The Extreme Ultraviolet Telescope(EUT),as a high-resolution imaging instrument,its optical performance measurement directly impacts on the image quality,so the imaging performance of aligned EUT is tested in this paper.The measuring method is to place a resolution test-target on the focal point of a collimator and the target is illuminated by a avisible light.Then,the visible light is becomed a parallel light to go through the EUT and image at focal plane on the CCD.By analysis the image,the resolution of the telescope can be obtained.The test results indicate that the angular resolution of the EUT is 1.22″(λ=570 nm),which is very close to the resolution limit(1.20″).After the computation,it concludes that the operation wavelength resolution is 0.32″,which can meet the requirement for design.

In order to measure the changes of optic characteristics for spacial materials before and after space radiation,a in situ optic measuring system with high precision and wide scanning wavelength was designed to measure feeble signal accurately buried in noise based on a virtual lock-in technique.The system structure,and measuring principle were introduced.Then,the G language Labview was used to develop a program to implement the virtual lock-in technique,and apply it in the system.After that,the light source intensity distribution and the system measuring repetitiveness were measured.Finally,proposed system was used to measure the transparence of a quartz glass.Compared with that of lambda950,the measured results show that the mean error between them is 0.95%.The experimental results indicate that the virtual lock-in technique not only restrains effectively the noise brought by interferential light and electronic equipment,but also simplifies the system structure;This in-situ measuring system by using such technique can accurately measure optic characteristics of spacial materials.

A novel method based on Least Square Support Vector Machine(LS-SVM) and genetic algorithm to select the temperature sensors of a Numerical Control(NC) machine tool was presented.The measurement points in a CNC lathe were grouped based on the thermal mode theory,Then,the genetic algorithm was used to determine the positions of optimum sensors.Finally,a thermal error regression model was established by the LS-SVM and a compensation model for the machine tool was given also.The results show that the novel method combined genetic algorithms and LS-SVM well avoids the correlation of the temperature sensors and ensures the accuracy of the model.In the experiments of the CNC lathe,the mean absolute percentage error of the LS-SVM model is 1.89% in axial direction and 2.04% in radial direction,it also can reduce costs and shorten modeling time for less temperature sensors.

Repeat sun-synchronous orbit is one of the most important orbits for earth observation missions,because the orbit can satisfy the requirements of illumination and temporal resolution.Q value is the connection between sun-synchronous orbit and repeat orbit,so Q value plays an important role in repeat sun-synchronous orbit design.In this paper,the characteristics of sun-synchronous orbit and repeat orbit are analyzed,and the relationship of Q value and repeat sun-synchronous orbit is established.With the analysis of repeat orbit characteristics,a Q value design method of repeat sun-synchronous orbit is derived,and an optimal orbit is selected using the method.The selected orbit shows that the orbital altitude is 502.59 km,repeat period is 21 days and mean revisit period is 5 days,which is an optimal repeat sun-synchronous orbit for the revisit characteristic within the range from 500 km to 510 km.It concludes that this method finds the key points of repeat sun-synchronous orbit design in theory,and provides a credible and easy method for complex orbit design tasks.

In order to resolve the problem that Laser Tracker System(LTS) can not continuously measure Corner Cube Retro-reflector(CCR) position in dynamic conditions for CCR's receivable angle is restricted,an angular automatic calibration mechanism for LTS is designed to measure the CCR positions in real time.The calibration mechanism mainly is consists of a precision circular guide-way and a angular automatic adjustment device,which can make the CCR track the laser beam emitted from LTS all the time.The designed calibration mechanism and Hexagon Globle9128 coordinate measuring machine are used to compensate the angular measuring error of LTS.The experimental results demonstrate that the maximum angular measuring error of LTS in measuring distance of 0.752 m is decreased sharply from 34.69 μm to 9.71 μm in horizontal direction and from 35.43 μm to 10.03 μm in vertical direction,which shows the LTS angular measuring accuracy is effectively improved.

In order to investigate the copper gap-filling in micro electroplating process with additive(N'N-diethylthiourea),the electrochemical behaviors of electrolyte were analysed by SEM,CVS and XRD,and the electrode dynamic parameters were studied by the Tafei equation.The results show that when N'N-diethylthiourea is used in micro electroplating copper process,an activation polarization is generated to improve the activation energy,the metal ion discharge rate is lowered from 2.221 4 mA/cm2 to about 0.076 mA/cm2.Therefore,the overpotential is increased,and the crystal nucleus molding speed on the electrode is accelerated;so that the crystal growth speed is decreased from 2.57 μm/min to about 0.17 μm /min,and the leveling ability is increased about 50%.Experiments show that the side effect is lowered effectively,which makes the copper ions get a good filling ability for micro-trenches.Furthermore,some micro trenches in the silicon wafer with the width of 10 μm and aspect ratio of 4:1 are filled by metal copper in micro electroplating process with N'N-diethylthiourea,the electroplating layer have no voids or seams.

In order to fabricate vitreous microfluidic components with round cross sections and microchannel networks with nude structure,a hot rheologic deformation puller for fabricating vitreous three-way microchannel was designed.In the forming process of the three-way microchannel,the tip of the bended glass capillary with the shape of "V" was softened by heating and then was pulled into three-way microchannel in cooling process.The reservoirs connecting the microchannels with conical pipes were also formed during the pulling process.The three-way microchannels with both equal or unequal inner diameters could be made by the puller.The temperature in the heating field,the inner diameter of microchannel and the distribution coefficient of vitreous material could be adjusted by heating time,pulling distance and micro-feed mechanism,respectively.A three-way microchannel with inner diameter of 67,32 and 20 μm was fabricated,its three ways were connected smoothly in the joint.For surface tension forming,the three-way microchannel shows a higher quality surface.Finally,a microchannel network was assembled with the one-dimension microchannels and the three-way microchannels successfully.

To resolve the problem of the gyro decoupling on two-axis and three-axis photo-electric platforms,and improve their stabilizations,the gyro coupling reason of two-axis photo-electric platform was analyzed,and a secant decoupling algorithm was given.With combination of the principle of flexible dynamic tuning gyroscope and the specific structure of photo-electric platform,the gyro coupling reason of tree-electric platform was discussed,and a sine decoupling algorithm and the ARM-based MCU decoupling hardware and software processing techniques were given.The experimental results indicate that the coupling amplitude value decreases from 22 mV to 3 mV,which means coupling degree decreases by 86.4%.It comes to the conclution that proposed algorithm improves the control performance and stability accuracy when the platform turns at a high speed.

In order to investigate the mechanism of acoustic landmines detection,an acoustic-to-seismic coupling based experimental system was developed,and a curve-fitting method was adopted to study the measured complex signals.Acoustic-to-seismic coupling theory was briefly described,and the resonance mechanism of soil-mine system was analyzed.Then,a loudspeaker and a geophone array were employed to induce and receive the ground vibration respectively.Finally,an average algorithm was used to fit the curves of ground vibration velocity and to explain experimental phenomenon.Experimental results indicate that the ratio of ground vibration velocity on-to-off mine is greater than 1 in wide frequency range and up to 24 in peak frequency.It proves that landmines can induce the distinct changes of ground vibration,and proposed experimental system and curve-fitting method are applicable for further study of acoustic landmines detection signals.

For acquiring the mean line of workpiece profile,a new kind of digital smoothing filter based on cubic generalized spline function was proposed.The generalized B spline functions were derived from the linear differential equation,and the digital smoothing filter consisting of a IIR filter and indirect generalized B spline transformation was deduced by combination with generalized B spline functions and variational principles.Then,the recursive algorithm of the filter was presented concretely,and the filter's universal property was testified.Finally,this filter process was applied to a workpiece to acquire mean line.The experimental results indicate that the filter has a better cut-off property than that of Gauss filter,and total calculation only requires 0.002 s for 11 200 data points by PC.The filter has advantages in adjustable smoothing property and zero phase transmission,which can overcome the drawbacks of low curvature and serious oscillations of traditional cubic spline interpolation.The recursive algorithm is efficient and simple.

According to the output performance of inchworm-type piezoelectric actuator,a novel close-loop control method based on the subsection experiential equation between actuating voltage and corresponding stepping motion displacement was proposed.Via the method,the control system could search for the optimal actuating voltage step by step so as to approach to the target position.A experiment on the closed-loop system of the inchworm-type piezoelectric rotary actuator was carried on.The experimental results show that the novel method can control the actuator for precision positioning in closed-loop mode,and the positioning error of the actuator system in close-loop mode is 0.18 μrad,which is improved by 14.4% as compared to that of the open-loop working mode.The control method has the advantages of high accuracy,quick response and so on.

A self-organized Local Linear Embedding(SO-LLE) algorithm is proposed.This new algorithm can not only determine the value of the nearest neighbor automatically and decrease the operation quantity,but also can gain a perfect approximation of face manifold.Theoretical analysis is given in details and experiments are carried on some dataset to verify the effeteness of this method.The experimental results on Yale and PIE face databases show that the proposed algorithm can increase the average recognition rate from 5% to 40%,which improves face recognition performance effectively.

In order to ensure the security of transmitted digital image effectively,an invertible chaotic 3D map based image encryption approach and its optimized algorithm are proposed based on the stretching and folding mechanisms of chaos.A gray-level image is depicted as a 3D matrix and mapped to a 2D image matrix according to proposed algorithm.Then,stretching and folding maps are used to permute the pixel positions of the 2D matrix.The stretching process is to transform the 2D image matrix to an array,and the folding process is to fold the array to another 2D image matrix with the same size as the original 2D matrix.Finally,the permuted 2D matrix is mapped back to a 3D matrix to obtain a cipher image.This proposed approach is invertible,it could be used for image encryption and decryption.The encryption and decryption algorithms are formulated,by which image encryption and decryption will do a large of calculation and consume a lot time for the image has large data quantity.In order to solve this problem,an optimized algorithm is proposed.Simulation results indicate that proposed optimized algorithm can increase image encryption speed,which is 3～4 times faster than that of the deduced formulas,proposed image encryption approach also can realize pixel permutation and substitution at the same time via one iteration map for the histogram information of the plain image,and can enhance the capability of resisting statistic attack.

The effect of pulsating magnetic fields on the transient outward K+ channel of rat cortical neurons was researched.This experiment used pulsating magnetic fields with frequency of 15 Hz and intensity of 1.4 mT to impose on acutely the isolated rat cortical neurons,and studied the properties of transient outward K+ channel using the whole-cell patch clamp technique.The experiment reveals that pulsating magnetic fields reduce the amplitude of IA,and significantly affect the activation process of IA.The half-activation voltage and the slope factor of the activation curve are changed by the pulsating magnetic fields' exposure from(13.25±2.22) mV to(30.98±4.11) mV(n=6,PP>0.05),respectively.The results show that pulsating magnetic fields can change the properties of transient outward K+ channel and can adjust the physiological functions of neurons by the pulsating magnetic fields.

A data collection method is proposed to support hybrid sink mode in wireless sensor networks where static and mobile sinks coexist.One or more mobile sinks cooperate with the static sinks to broadcast status update periodically only to a subset of sensor adjacent nodes,Then,the sensor nodes send or transmit data packets to the nearest sink in hop counts.The correct data transmission between the mobile sinks and the last hop sensor node is guaranteed by using ACK-RETRANSMIT method.Simulation result shows that the network lifetime has been improved with the increase of the number of mobile sinks.When the ratio of the mobile sinks to the sensor nodes is above 7.5%,more than 600% lifetime improvement can be achieved.When the load of network is high,the maximum success rate of data transmission reaches to 91%,which has increased by 8.3 times as compared with that of the network in only one static sink,meanwhile,data transmission delay remains at a low level.

To solve the problems on the important ultrasonic noise and the quality degradation of data in ultrasonic imaging,a novel image enhancement algorithm by calculating the Non-stationary Degree(NSD) of the image is proposed,which highlights the rapid changes of image to realize the image enhancement.The simulation results are given for different SNR levels of input signal by applying the NSD algorithm and the Deriche operator respectively.Two successful actual applications for a rabbit's liver and a 12-week infant,are provided.The research results show that proposed algorithm is insensitive to the noise.Compared with the image without processing,it is obvious that the enhancement of the ultrasonic stereovision image is effective.

To measure the ultra low current in a quadrupole mass spectrometer(QMS) by a preamplifier,low noise and wide bandwidth is the key to this system for its current is from 10-11 A to 10-6 A and varies quickly.So a Non-inverting Input Composite Preamplifier(NIICP) is designed,and the effect of the open-loop gain on optimizings the frequency responses of signal and noise is analyzed.NIICP can optimize the openloop gain to reduce intrinsic noise in the lower magnification and to recover high frequency signal damped down by the low pass transfer function of large feedback resistance and parasitic capacitance in the higher magnification.Compared with Inverting Input Composite Preamplifier(IICP) in the commercial QMS,the NIICP can reduce noise by 5%～30% in the 106～107 V/A magnification,and can expand signal bandwidth by 1.8 times and increase MS peak amplitude by 10%～15% in the 108 V/A magnification.Experimental results show that the wide-band low-noise NIICP improves the signal-to-noise radio of MS peak in large dynamic range of 106～108 V/A.

In combination with Wavelet Transform(WT),Two-dimensional Principal Component Analysis(2DPCA) and Ellipsoidal Basis Function(EBF),a fingerprint recognition algorithm based on WT,2DPCA and EBF neural network(EBFNN) is proposed.Original images are decomposed into high-frequency and low-frequency components with WT,and horizontal and vertical high-frequency components are ignored,so the prime features of original images can be obtained;then,the projected features are solved by 2DPCA;finally,fingerprint recognition can be realized by EBFNN.The algorithm combines the optimization of 2DPCA and the adaptability of EBFNN and achieves the accurate recognition rate of 91.4%.The experimental results based on FVC2000 verify that proposed algorithm has higher recognition rate than that of WT-PNN and WT-2DPCA-RBF.

Due to the universality of Cellular Automaton(CA) model,many applications have been found in traditional cryptography and image processing.Combined with Cellular Automaton Transform (CAT) compression technologies,an image encryption approach with enhanced security and more larger key space at least 2H×2512×233 554 432 is proposed based on sequential CA.The simulation results show that the proposed scheme has the advantages of convenient realization,avalanche effect,confusion and diffusion properties,and low cost,which is suitable for optical realization based on the characteristics of simple parallelism structure of sequential CA,very large security key space and fast encryption speed.

In order to obtain optimization threshold for the centroid algorithm,several kinds of influencing factors,such as signal-to-noise ratio,discrete gathering of star point and so on,are studied.Based on theoretical computation,a function is established to depict the relationship between the luminosity of star spot and the threshold with optimization centroid accuracy.Then,by combining the function with the simulation results of discrete gathering of star spot,a method of obtaining the optimization threshold based on luminosity of star spot is proposed.Based on the relationship between optimization threshold and luminosity of the spot,a curve model is established,and the optimization threshold is educed by interpolation.The threshold estimation accuracy of the method is shown,and the centroid accuracy is also compared with those of various thresholding algorithms.The experimental results indicate that the luminosity estimation error of the optimization threshold is less than 3 under different SNR conditions of the camera system,and the centroid locating accuracy by proposed method has raised by 1/100～1/10 pixel level as compared with those of other methods,which can satisfy the requirement of dynamic threshold optimization of centroid algorithm.