To meet the need of 1 m Seya-Namioka monochromator for gratings in the combustion and flame endstation at National Synchrotron Radiation Laboratory( NSRL),1 200 lp /mm Laminar gratings were succesfully fabricated with holographic lithography and ion beam etching.The duty cycle of a resist grating mask was adjusted by resist-ashing, then the resist grating mask was etched in the range of theory design error and its pattern was transfered on a substrate by the scanning ion beam etching method. A gold film with a thickness of 40 nm was coated by ion beam sputtering and an aluminium film with a thickness of 60 nm by evaporating on gratings, respectively, after the rest resist was removed and cleaned. The microstructures of gratings were characterized by an Atomic Force Microscopy(AFM), and results show that the groove depth is 40 nm and duty cycle is 0.45 for the gratings. Synchrotron radiation wavelength scanning were also performed at the exit slit of the combustion and flame endstation, the results indicate that the photon intensity of Al-coated grating is better than that of Au-coated one which agrees well with the calculated results.The Au-coated grating has been used for experimental investigation for 3 years, futhermore its lifetime has been larger than that of the replica grating.

Based on a wavelength scanning technique and the characteristics of ultra-narrow-linewidth laser, a kind of cavity enhanced absorption gas detection system with high sensitivity is designed. A semiconductor laser with ultra-narrow-linewidth and tunable wavelength is used as a light source, and an optical cavity which consisted of two mirrors with high reflectivity is also used as an absorption cell in this system. The laser frequency overlaps with one of cavity modes via scanning the cavity length and the input laser is switched off by using the laser detuning technique.Then, the ring-down detection of acetylene with low concentration is achieved. With the cavity enhanced absorption technology, spectra of acetylene near the region of 6 518.824 cm-1 and laser ring-down time at different concentrations are measured and also analyzed. The experiments for concentration measurement indicate that the relative error is less than 2.5%, relevance coefficient is 0.999 and the lowest detection limit is 2×10-6. By filling a volume of acetylene into the gas cell gradually, the system dynamic response time is less than 10 s. Obtained results show that the system has advantages of good accuracy, high sensitivity and quick response, which can be used in the fault gas on-line monitoring for power transformers in real time.

To improve the wire bonding and homogeneous injecting currents, this paper proposes a Vertical Cavity Surface-emitting Laser (VCSEL) array with a non-closed structure. In this structure, a non-closed groove is etched to form a mesa for simplifying the processing and reducing the damage of devices. The 850 nm VCSEL arrays (including 2×2,3×3 and 4×4 array devices) with the non-closed type top emission structures are tested and analyzed. Results show that the continuous output powers are up to 80, 140 and 480 mW at room temperature, threshold currents are 0.15, 0.25 and 0.4 A, and the far-field divergence angles in parallel and vertical directions are 9° and 9.6°,13.5°and 14.4°,15° and 14.4°, respectively. In the context of a pulse width of 50 μs and a pulse repetition rate of 100 Hz, the maximum output powers are 90, 318 and 1 279 mW and the threshold currents are 0.2, 0.5 and 0.7 A, respectively. The power curves of chips are tested before and after packaged. The result shows that the heat saturation current of the chip after packaged is much higher than that of the chip before packaged, which suggests that good packaging can improve the cooling efficiency and reduce the effect of internal heating on the device performance.

The Fractionally Spaced Equalization (FSE) technology is investigated to mitigate the effects of Inter-symbol Interference (ISI) on the indoor Visible Light Communication (VLC) systems. Based on the propagation properties and signal modulation characteristics of indoor VLC systems, a link model for the indoor VLC communications is described, then the FSE method is proposed by using the link model. Furthermore, based on the minimum Mean Square Error (MSE) criterion, the FSE is optimized. Finally, the performance of the proposed FSE theory is evaluated by computer simulation. Experimental results indicate that the performance of T/2-fractionally spaced equalizer has improved 1-2 dB than that of the symbol interval equalizer in the same Bit Error Ratio(BER). It can eliminate the effects of the ISI induced by the multipath effectively, and can improve the reliability of signal transmission.

With the aim to lower the surface roughnesses of etched samples and the etching threshold fluence of a direct laser irradiation source and to raise the yield of micro-optical elements, a laser induced backside dry etching technique with the assistance of a solid medium as the absorbed layer was proposed to fabricate transparent dielectric materials by direct laser irradiation. By using an alumina ceramic wafer (Al2O3 in 95%, surface roughness less than 500 nm) as the absorber, a 2D transmittance grating in a micro size binary Diffractive Optical Element (DOE) was fabricated on the fused silica with a thickness of 3.2 mm by the 1 064 nm Ytterbium Doped Fiber (YDF) laser. And then, the etching parameter curves were fitted and the effect of laser energy density on the parameters was discussed. Finally, the diffraction patterns of micro structure were observed to examine the features of binary DOE. According to the measured results, the binary transmittance grating shows its grating constant in 25 μm,the depth of trench in 4.2 μm, and the RMS roughness of the trench bottom below 40 nm. Furthermore, the etching threshold influence is estimated below 7.66 J/cm2. These results are much lower than those of the etching without any absorbers.

A set of force actuator was designed and tested for the active support system with a thin primary mirror to control the mirror surface error all the time. Conventional mechanisms for the force actuator with high accuracy and fine stability in engineering applications were studied. Then, based on the requirements of actuality and implementation in the thin primary mirror experiment system, a set of force actuator was designed and calculated in detail. A stepper motor integrated with a harmonic reducer was used to actuate the mechanism and precision screw to transmit, and an S-type Loadcell was taken as the force sensor to realize the feedback of force output. Finally, the mechanism was tested to validate the design feasibility through the opened loop and closed loop controls. Experimental results indicate that travel range of the force actuator is 0-10 mm, force range is -100-100 N and the accuracy is better than 0.05 N, which satisfies the demands of the force actuator for long range, high accuracy, fine output and stability. The actuator can be used in the support system of active optics, and is also a useful mechanism to other fine-tune structures.

The 0Cr18Ni9 stainless steel film with a thickness of 20 μm was ablated and cut in a micro-machining. Firstly, the ablation threshold of 0Cr18Ni9 stainless steel film and the beam waist radius of a femtosecond laser were obtained based on the functional relationship between ablation area and laser energy. Then, the metallographic phase and electrical resistivity of the stainless steel film ablated by the femtosecond laser were researched in detail to observe the heat effect caused by the laser. Finally, in order to research the effect of femtosecond laser ablation on the phase composition of the stainless steel film, specimens were tested by X-ray Diffractometry(XRD). Experimental results demonstrate that the waist radius of femtosecond laser is 10.416 μm and the ablation threshold of the stainless steel film with a thickness of 20 μm is 0.455 J/cm2. The metallographic phase of specimens cut by the femtosecond laser is in a recovery process, and its metallographic phase is not changed. The phase composition of stainless steel film is not affected by the femtosecond laser cut, however, the content of phase has been changed.

As traditional grating-based spectrometers can not offer fast apertures and small volume systems synchronously, this paper designs a convex grating by using holography and develops an imaging spectrometer with the convex grating as a core element. It acquires the image of target by pushbroom scanning, then obtains the data cubes of the imaging spectra. The image spectrometer has a space resolution of 2.4 nm, a spectral line bend of 0.1%, the chromatic distortion of 0.6%, and a volume of 209 mm×199 mm×110 mm. The principle of imaging spectrometer with convex grating is analyzed and its constructs are described. Then, a spectral experiment in the laboratory is carried out and the spectra of extraventricular flowers are tested. The result demonstrates that the real spectral resolution is 2.1 nm, line bend is 0.09% and the chromatic distortion is 0.6%. These data satisfy the design requirement, and obtained spectra are ideal results.

To enlarge the grain size and decrease the effect of the defect between the grain boundaries on a p-Si film transistor, a phase modulated excimer laser crystallization technique is used to fabricate the uniform p-Si film with a large grain size. First, an energy window for super lateral growth is determined by measuring the grain size of p-Si film fabricated with different laser energy intensities. Then, the spatial distribution of the input laser is modulated by a phase mask with a period of 1 073 nm and an artificially controlled lateral temperature gradient is induced on the a-Si film, which leads the a-Si to be melten and crystallized into p-Si grains by super lateral growth. Finally, the characteristics of the prepared p-Si film are measured and compared with those of the a-Si film and the p-Si film fabricated by super lateral growth technique. The results show that the grain size of the p-Si film is 228.24 nm, which is ten times of that fabricated by super lateral growth under the same processing parameters; the electrical resistivity of the prepared sample is 18.9 Ω·m, which is lower by an order magnitude than that prepared by super lateral growth. Furthermore, the distribution of the grain is more uniform than that fabricated by other techniques. The reported technique can increase the electrical characteristics of the p-Si film greatly and is suitable for the fabrication of high quality p-Si devices.

According to the requirements of Light Emitting Diodes (LEDs) for consistency and reliability, when they were taken as a light source in spaceborne calibration, a screening experiment scheme using multiple LEDs as spaceborne calibration light source was proposed, and a screening device was developed. In order to test the working stability and the consistency of luminescence for the LEDs, this device could monitor and measure the luminous status of multiple LEDs automatically at the same time. Using this device, 72 white LEDs which were randomly selected were tested. According to the analysis and calculation of the experimental data, the result shows that the change of the relative intensity is about 1.5% after those white LEDs are lighting for 900 h. Finally, 10 desirable LEDs are screened depending on the requirements of the light source.

An electromagnetic force device is designed to realize the precise force control in the process of fusion stretching of ultra-thin fibers. First, the electromagnetic force with different coil parameters is calculated based on finite element method, and the relationship between electromagnetic force and coil parameters is derived. With the demand of stretching system performance and the dimension restrictions of the coil frame, constraint equations and objective functions are established to optimize the solution to machine the electromagnetic coil. Then the control circuit for electromagnetic force is designed to regulate the current in the coil and achieve the fine adjustment of electromagnetic force. Finally, experiments of electromagnetic force device are performed. The results indicate that the range of stretching force for the optical fibers can achieve 26.073 mN and the measuring resolution is 7.473 μN. It satisfies the demand of the fusion stretching of ultra-thin fibers.

This paper studies the influencing factors on the electromagnetic shielding of an inductive metallic mesh when the infrared transmission reduction of optical window caused by the mesh is less than 5％. The infrared transmission formulas of the inductive mesh are summarized, then, the electromagnetic field integral equation for the periodic structure of metallic mesh is built based on Galerkin's method in the spectral domain. The transmission and reflection coefficients are derived by the Moment of Method(MOM). After that, the electromagnetic shield effectiveness of the mesh is calculated and the effects of different line widths, periods, substrate materials, substrate thicknesses on the electromagnetic shielding of the transparent conductive optical window (with metallic mesh) are calculated and analyzed. Finally, metallic mesh samples with a line width of 12 μm, a period of 360 μm×360 μm and the surface resistance of 13 Ω,25 Ω are fabricated on a ZnS substrate by the laser direct writing figure, vacuum coating, et al.. The electromagnetic shield effectiveness from 2 GHz to 18 GHz is tested by a free space method. Results indicate that the average electromagnetic shield effectiveness is more than 20 dB when the infrared transmission reduction caused by the metallic mesh is less than 2％ in 8 -10 μm. Experiments indicate that the optical and electrical performance of metallic mesh film is inconsistent .The main solution method is to select a thinner line width and smaller period and to reduce the surface resistance value of the mesh as low as possible at the same time.

The dynamic model and inverse model of a piezoceramic actuator were proposed to improve its control precision in a precision position system. According to the Weierstrass approximation theorem, the polynomials f(·) and g(·) in the Duhem function was developed, and the dynamic modeling of nonlinear parameters of the piezoceramic actuator was given by using recursive least squares to identify the model parameters and polynomial coefficients in the Duhem model. Then, an inverse dynamic modeling of the piezoceramic actuator was established based on identified results to simplify the unknown parameter computation process greatly. Finally, the dynamic inverse compensation was incorporated in a closed-loop PID controller to control the piezoceramic actuator. Experimental results indicate that the maximum absolute error with the inverse compensation is less than 0.8 μm and that with the inverse compensation and PID is less than 40 nm in an amplitude range of 200 μm. The experimental result shows that the proposed identification scheme has improved the nonlinear characteristic of the piezoceramic actuator effectively.

It is necessary to determine camera postures in inertial coordinate systems, when the mapping camera is used in a photogrammetry for a transport stereo mapping satellite. In determining the camera postures, the postures of star sensor measuring system in the inertial coordinate system were measured by star sensors, then, the postures of the mapping camera measuring system in the inertial coordinate system were obtained by the transition matrix between star sensor measuring system and the cubic prism in star sensor and the transition matrix between the two cubic prisms in the star sensor and mapping camera. This paper introduces the definition of coordinate systems. By using the measuring systems of four theodolites, we establish coordinate systems on two cubic prisms respectively and a transition matrix for the coordinate systems. The calibration method of two coordinate systems is presented. The measuring results from many experiments show that the maximum calibration error is 1.011 6″,which is better than 2″(1σ) and satisfies the precision requirement of the stereoscopic mapping.

As the sensitivity of a piezoelectric mass sensor directly depends on the structural frequency variation induced by the mass added. This paper proposes a structure designing method for improving the measuring sensitivity of the mass sensor by using a structure which consists of a symmetrical trough (I-shaped cross-section) cantilever and piezoelectric films. Then, we design and fabricate a novel piezoelectric resonance micro-mass sensor. Considering the influences of the section shape, natural frequency and the vibration mode on the sensitivity, an analytical model is established for analyzing the frequency variation caused by the micro particles. With the same geometric parameters, the simulation and experiments are performed for the sensors with the I-shaped cross-section cantilever and the rectangular section cantilever. It shows that the first order natural frequency of the I-shape cross-section cantilever is 1 851 Hz, and that of the rectangular section cantilever is 1 610 Hz. Moreover, corresponding sensor sensitivities are 3.12×104 and 1.5×104 Hz/g, respectively, and the former is twice of the latter. The method is feasible and effective for improving the measuring precision of mass sensors.

A universal interferometer is proposed to realize the eyeballed evaluation and accurate measurement of many kinds of optical elements with different structures and shapes, such as lenses and crystals. In the interferometer, a pair of optical wedges are used to adjust the fringe periods dynamically to implement the measurement in real time. The interferometer takes two Jamin glass plates as a main structure, between which two pairs of optical wedges are laid. All wedges are completely identical in sizes and can be revolved along the axis. The relationship between the relative angle of two wedges and the fringe period is established, and then a real time testing experiment of domain reverse in an optical crystal is introduced to test the performance of the interferometer. The experimental result shows that the measurement accuracy is about 0.2 rad in phase (λ/30), and the adjustment range of fringe period (white & black fringe) is from 0.105 to 5.993 mm/pair. In the interferometer, the relative angle between the two wedges along the same optical axis can be adjusted from 0° to 179°. The interferometer is charaterized by the symmetrical optical structure, so that it has high stability and can endure strong noise and vibration. It can modulate the fringe period in real time to meet requirements of various optical structures and can reach high measurement accuracy.

A Continuous Multi-point Forming (CMPF) method was researched to realize high efficiency and flexible manufacturing for rotary surfaces. The principle of CMPF was described and its characteristics were analyzed by comparing with the traditional rotary surface manufacturing method. By taking a disc-shape suface for an example, the Finite Element Model(FEA) was established, and equivalent stress and plastic strain distributions were analyzed. By simulation results and principal strain vectors, the wrinkling was discussed. Finally, the CMPF equipment was developed, and experiments were performed. Results indicate that the equivalent stresses in the region of center fixing and a flexible roller exceed the yield stress, and their maximum value is 213 MPa; the maximum plastic strain is generated in the center region, and its value is 0.07; the plastic strain in the region of flexible roller takes the second place, and its value is 0.03. Furthermore, shell elements in the wrinkling region generate the compress deformation in a tangent direction in 605-615 mm which accord with the simulation results. It concludes that the simulation results of stress field, strain field and wrinkling are in good agreement with practical situations and the CMPF equipment can manufacture good rotary surfaces.

A calibration technique for the three-dimensional measurement system of gas-liquid two phase flow based on a virtual stereo vision was researched. Firstly, a single high-speed camera and two groups of mirrors were used to construct the virtual stereo vision system and to establish a perspective transformation model for the high-speed camera and a virtual stereo vision three-dimensional measurement model. Then, the camera and the virtual stereo vision sensor in the virtual stereo vision system were calibrated. Finally, the distribution of the bubble in the water was simulated by a target standard ball and its space distance was regard as the measurement standards. Furthermore, the impacts of different calibration methods on the accuracy of three-dimensional reconstruction were compared. Experimental results indicate that the accuracy of the three-dimensional reconstruction is the best when the calibration reference is placed in a water tank and the left and right virtual cameras and sensors are calibrated, respectively. The absolute error and the relative error of the measurement distance are better than 0.13 mm and 0.49%, respectively. It suggests that the impact of the light splitting path and tube wall refraction on the calibration accuracy should be taken full account of in the calibration of the gas-liquid two-phase flow three-dimensional measurement system based on the virtual stereo vision system.

In the Off Pump Canary Artery Bypass Graphed (CABG) surgery, a robot is usually used to dynamically track the Point of Interest (POI) on the beating heart. To achieve better tracking performance and cancel the relative motion of the heart and the assisted robotics,an adaptive Auto-regessive (AR) linear model of heart motion was built to convert the complicated heart motion tracking problem to a dynamic model following problem. The Kalman filter was used to estimate the motion state of the assisted system and the linear quadratic optimal tracking theory was used to implement the model-following method. The results show that the model-following method which provides estimated future reference enables the robot to improve the relative motion cancellation ability by 30% and decrease the tracking error by 0.25 mm. The comparison tracking result on 3D test bed Phantom robotics is reported, which proves that the model-following method enhances the ability of dynamic relative motion cancellation during the CABG sugary.

Several kinds of current boundary classification methods based on hyperplane, hypersphere or ellipsoid were studied, and a novel classification method called Maximum-margin Fuzzy Classifier (MFC) was proposed by using a space point as the classification criterion. By the proposed method, a fuzzy classified point c was chosen in the pattern space firstly, which should be as close to two classes as possible. Moreover, the angle between the two classes should be also as large as possible. Then, the testing points could be classified in terms of the maximum angular margin between c and all the training points. Finally, the applications of the MFC were popularized from two-class classification to one-class classification according to the kernel dual of MFC equivalent to the Minimum Enclosed Ball (MEB). Comparative experiments with current classification methods verify that the MFC has good classification performance and noise resistance ability and its classification accuracy has been reached 98.9%.

A recording method to suppress the zero-order diffraction term in an off-axial digital hologram and to improve its effective bandwidth is proposed. The method records the hologram on the condition that the reference intensity is more stronger than that of the object. Then, the recorded hologram is normalized by the recorded reference intensity beforehand. The natural logarithm operation and Hilbert transform are applied to the normal hologram to suppress the zero-order diffraction and the object complex field is recovered by an exponential operation. The theory of the method is analyzed. It shows that the logarithm operation can suppress the zero-order term and the Hilbert transform can filter the desired term. Finally, the experiments are performed to validate the method. The results show that the method can suppress the zero-order diffraction term efficiently even in the case of the strong spectral overlap between the zero-order term and the diffraction orders. The influence of the intensity ratio of the light of reference to that of object on the suppression efficient of zero-order diffraction term is analyzed. It shows that the zero-order term can be efficiently suppressed through the proposed method when the intensity ratio is above 5.

An image test system using a CCD and a standard target was established to measure the stabilization accuracy of a gun control system. The principle and structures of the system were introduced. and its key algorithms such as image pretreatment, template matching, the orientation of target cross center and system calibration were investigated. In order to remove the disturbance of the reticle of graduation board on the target cross, the algorithm of the switching median filter was presented. After choosing the grayscale transformation to improve the image contrast, the algorithms for fabricating and matching the template were analyzed and chosen. Then, the orientation algorithms of pixel and subpixel for target cross center were presented. Finally, the angle resolution of the system pixel was confirmed using a calibration method based on active vision. The result indicates that the test accuracy has been better than 0.07 mil, which meets the practical test requirement and implements the stabilization accuracy measurement of modern fire-control systems with stabilized guns in firing ranges.

As the common dynamic test method for Micro-electro-mechanical System (MEMS) is limited by its tiny feature sizes, this paper proposes a dynamic test system for micro machined gyroscope based on high speed photography after introducing three kinds of dynamic test methods. The hardware of the system and the function of the software are introduced. Then, it explains the key algorithm of the dynamic test system. It uses the Otsu algorithm to segment the image, and takes the region of Interest (ROI) and moment invariants to obtain the vibration curve. After spectrum analysis, the vibration characteristic of the micro machined gyroscope can be gotten. Experimental results indicate that the precision of the amplitude is 0.1 μm and the repeatability and consistency of the frequency domain is satisfactory. The dynamic test method shows higher precision, rapid speed, as well as strong anti-jamming and stabilization. It reduces the amount of calculation and improves the operation speed for the system.

A segmentation and detection method for small infrared targets is proposed to improve the accuracy of target detection. Aiming at the characters of an infrared image, an improved background perception algorithm is used to suppress backgrounds for increasing the target detection probability, and the bound set is constructed to limit the gray level histogram into a Region of Interest (ROI) to reduce the interference of background. Then, target detection is achieved through image segmentation using the maximum entropy of a 2D bound histogram. Furthermore, the fast recurring algorithm is applied to the proposed algorithm for accelerating the running speed of the segmentation algorithm. Experiment results show that proposed background suppression algorithm has the better performance in background suppression and can improve the signal to Noise Ratio (SNR) of the image. Moreover, the segmentation algorithm shows a better effectiveness in target segmentation and detection, and its candidate target is separated more accurate with less false alarm points and running speed has improved by 91%. Through post-process for the image, most of the false alarm points are eliminated, which provides powerful guarantee for the subsequent accurate detection.

To realize automatic target detection, an algorithm is proposed to detect small visible optical space targets against low SNR conditions. Firstly, the single-frame image background is segmented, and the segmentation coefficient is determined by a Constant False Alarm Ratio (CFAR) method. Then, a feature space is formed based on structural stability of the star, and classification criterion function is constructed for the distance feature space. Furthermore, candidate targets are extracted by using the iterative optimization distance classification method. Finally, small visible optical space targets are detected by trajectory association based on the continuity of target motion. In addition, an evaluation method combined with single frame detection probability, single frame false alarm probability and sequence detection probability is proposed. Experimental results indicate that the detection probability of sequence is more than 96.02%, and the false alarm probability is less than 4.4% when the SNR≤3. It concludes that the method can promote the detection probability against low SNR conditions significantly, and can remove the false alarm effectively.

An inspection algorithm based on cascade classifiers is presented for detecting small moving foreign substances with low Signal and Noise Ratio (SNR) and low contrast in sequential images. The algorithm obtains three features of absolute difference, local difference contrast and neighborhood correlation from the sequential images of an ampoule. Each feature corresponds to a classifier, and small foreign substances are inspected by using three-layer cascade classifiers. The first layer corresponds to a traditional frame differencing method, which is used to remove the background and detect the large moving foreign substances. The next two layers are used to inspect small foreign substances and remove the noises generated by optical flow and the stain of bottle. Experiment results show that compared with the traditional frame differencing method, this algorithm has higher detection precision and higher anti-interference ability in inspecting small substances with the interference of a complex background, and the detection rate of small foreign substance is 99.3%. This algorithm can meet the requirement of real-time detection of ampoules for medicine production.

In consideration of the serious size distribution broadening and tailing phenomena come from improper iterative steps in inversing laser particle size distribution, this paper presents a method to select the iterative step for Projection algorithms to improve measurement accuracy. Based on the similarity theory, this method obtains the similarity between the row vector of light energy distribution and the course quantity of light energy coefficient matrix. Then, it gets the predict distribution of particle size after filtering and normalized processing. Furthermore, according to the predicted result, the iteration step can be calculated. The measurement results for national standard particles GWB(E)120046 show that the measuring errors for D50, D10, and D90 are -0.6%, -1.1% and -0.6%, respectively, and the light energy logarithm error of mix standard particles GWB(E)120041 and GWB(E)120049 is 2.167. Results demonstrate that the algorithm can effectively improve noise immunity, measurement accuracy and resolution.

For insufficient sampling data existing in the practical application of Photoacoustic Tomography (PAT), a reconstruction method based on total variation method was proposed for photoacoustic imaging to solve this problem. The residual between the real signals and the simulated ones from the reconstructed image was calculated for the update image, and the iteration was implemented to obtain the reconstruction image. During the process of iteration, the total variation method in the compressed sensing was utilized to obtain the image with the smallest total variation value by the gradient descent method. Through the numerical simulation, the image reconstruction in the case of insufficient sampling data was accomplished. The results demonstrate that the reconstruction method based on the total variation has better performance as compared with the filtered back-projection method, deconvolution reconstruction method and algebraic reconstruction method. For the 30 sampling points, the peak signal-noise ratio of the reconstructed image is 30.98, 22.09, 8.35 dB higher than those reconstructed by tree kinds of other methods metioned above, respectively. The result of in vitro experiment also shows that this method is more effective for noise suppression.