The coupled wave equations of volume holographic grating in photorefractive single-axis crystal in the discretional direction of the optical axis are derived in terms of proper coordinate transformation and Kogelnik's one-dimensional coupled wave theory. The influences of the optic axis on the isotropic and anisotropic diffraction properties of volume grating in LiNbO3 crystal are investigated. It is found that when the geometric parameters of the gratings are given, the diffraction efficiency of the isotropic and anisotropic Bragg diffractions can be maximized by properly choosing the direction of optical axis of LiNbO3 crystal. And the approximately proper direction of the crystal axis is given when the corresponding diffraction efficiency is maximum.All the theoretical results are very useful for the optimum design and wide applications of many volume holographic elements based on volume grating structures.

Based on a triple-clad numerical model, a long-period fiber grating chemical sensor coated with the sol-gel gas-sensing films on the cladding of the fiber grating region is studied. The relation between the sensitivity Sn and the thin film optical parameters (thickness h3 and refractive index n3), and the fiber grating parameters (the grating period, the core index modulation and the grating length) is analyzed. By using optimization method, the optimal optical parameters of thin film and fiber grating parameters are obtained to make the sensor highly sensitive. Numerical simulation shows that the sensitivity of this scheme to refractive index of the films is 10-8. In the end a long-period fiber grating sensor sensitive to C2H5OH is fabricated, which proves the necessity of the structure optimization of the sensor.

Based on the analysis of implementation method of NRZ, RZ and CSRZ formats, the transmission performance of the 10 Gb/s system with these different formats mentioned above was compared by numerical simulation, respectively. In this simulated system the EDFA was used as the solo amplifier and the CFBG was served as the solo dispersion compensator. A practical 8×10 Gb/s transmission experiment with CSRZ formats in straight line was also conducted and only 2.5 dB power penalty was cost after 2560 km transmission with error-free code. The experimental results show CSRZ will help to optimize the performance of the CFBG-based system greatly.

Based on the analyses of environmental satellite multispectral images, a new compression algorithm is proposed based on the three-dimentional set partitioning in hierarchical trees (3D-SPIHT) and region of interest (ROI) coding. To reduce the redundancies in the spatial and spectral domain and decrease the reconstructed spectrum error, a three-dimentional discrete wavelet transformation (3D-DWT) which combines two wavelet bases in the spectral domain is carried out. Then, a method which unites partial 3D-SPIHT with lifting the wavelet coefficients of ROI coding algorithm, is adopted to allocate the coding rate in reason according to the different significance of the reconstructed spectrum in the spatial orientation tree wavelet coefficient. The new algorithm on the probability of the significant coefficients in each bit plane to select one of three different coding modes, which enhances the coding efficiency. Finally, the experimental results show that the proposed algorithm achieves improved performance over the conventional algorithm. With the 8∶1 compression ratio, the algorithm satisfies the requirement of the satellite multispectral image system.

A new algorithm is proposed, through Fourier transformation projection theory using frequency-domain analysis, to detect local linear cracks on the freeway surface. And valid results are obtained. The steps of discrete ridgelet algorithm and the ridgelet transform results of the standard image are represented in detail. The basic condition of reconstructing the original image through Radon transformation is proposed. The method is used to detect actual complex surface, and denoise background noise combined with the histogram equilibrium algorithm. Estimated sample threshold is used to treat the level coefficients of ridgelet transform and get rid of the Random noise. The signal-to-noise ratio of ridgelet transform is better than that of the two-dimensional wavelet transform (greater than 20 dB in the state of low-frequency) and wavelet transform plus Weiner filtering (greater than 3 dB on the average) through counting the different reconstruction coefficients. The local linear crack is picked up through binary image and the resolution factor is within 2 mm.

Image fusion at pixel level without precise registration always causes wrong colors and other problem. Classification-based fusion scheme can effectively eliminate the incorrect color at the edge of the objective. But the traditional per-pixel classification results in the well-known salt and pepper effect. The only way to smooth the image is to use filters, and it would impact on the result of fusion. A method consisted of a sequential application of segmentation, classification and fusion techniques is proposed. Firstly, the object-oriented classification was used. And it can eliminate the influence of salt and pepper effect. Subsequently, according to the restriction of the precise classification result, adjusting the multi-spectral image, then the fusion by using Hue-Saturation-Value (HSV) color transformation. Finally, this merging technique was compared to other merging methods. The quantitative comparison result showed that it was not only provided a better visual representation but also had higher value in the variance and entropy.

A new digital watermark technique based on digital holography and discrete cosine transform (DCT) is presented. The digital holographic watermark is embedded into an original image. Because the hologram can't be destroyed, the digital hologram watermark is constitutionally robust to cropping. It can embed more information than traditional methods. The digital hologram is obtained by the means of Fourier transform holography. In order to enhance its robustness to lossy compression, the watermark is embedded by using JPEG model and decomposing DCT coefficients. The experimental results indicate that the proposed algorithm is robust to many image operations, such as JPEG lossy compression and image cropping. And the application of security lock provides a good security. So this algorithm can be used to protect the digital multi-media's copyright efficiently.

Scene matching real-time image synthesis plays an important role in simulation of aircraft navigation and performance evaluation of scene matching algorithm. The factors affecting low-light level (LLL) imaging including the change of illumination, meteorology, the characteristics of ICCD camera and the ground spectral reflectance are investigated. Mathematical models of LLL scene simulation are established, including ground reflectance model, ICCD camera model, atmosphere transmission model and noise model. The contrast difference between daytime image and LLL image is studied and a LLL image simulation scheme is proposed. Then a whole real-time LLL image matching synthesis system oriented to ICCD camera is developed using the computing kernel of LOWTRAN7. Simulation result shows this scheme is valid.

It is known that moiré tomography is applicable in widely changed flow fields and not sensitive to heavy vibrative conditions. So a new algebraic iterative algorithm developed from the essence of moiré deflectometry is presented to reconstruct three-dimensional flow fields. The algorithm is demonstrated by mapping a double-hump simulated temperature field by projections with 6 directions, 12 directions and 12 directions with Gaussian error added separately. Under same conditions, two other available tomography reconstruction algorithms, the filter back-projection algorithm and the algebraic reconstruction algorithm with integral operation along the deflectometry-projection data, are used to reconstruct the field. A simulation with an opaque is also made with the new algorithm. The reconstruction results show the new algorithm not only has strong ability of noise immunity, but also can handle reconstructions with incomplete data quite well.

A two-photon excited fluorescence lifetime microscopic imaging system based on a newly-developed high repetition rate picosecond streak camera is presented. The high repetition rate picosecond streak camera is the emphasis for introduction. In order to increase the measurable sweeping range and maintain high temporal resolution as well, and in order to realize high temporally resolved sampling measurement in two dimensions to improve sampling speed and exploit effectively the advantages of the streak camera, authors specially designed and developed a streak image tube with large area and high temporal resolution and a ramp voltage streak circuit with repetition rate up to 1 MHz. The streak camera based on the key components above mentioned is characterized with high repetition rate, tunable sweep rate, high temporal resolution, large effective area of work, low nonlinearity and small triggering jitter. A Ti:sapphire laser provides mode-locked, ultrafast femtosecond pulses with a fundamental frequency of 76 MHz which is reduced to 1 MHz with a pulse picker. The temporal resolution, sweep rate and nonlinearity are calibrated with a tunable delay generator and etalons. Temporal resolution of the system is 6.5 ps, nonlinearity is 2.60%, and the measurable time range is from ten picoseconds to tens of nanoseconds. Streak images of Rhodamine 6G and Coumarin 314 are obtained and the results agree well with those reported in the literatures.

The quarter-wave plate is the key device in the technology of polarized light. So it is important to determinate the retardation of the wave plate. A phase-modulated ellipsometry is described. By modulating the state of polatrzation, intensity signal changing with time is generated from the detector. The delay of the wave plate is obtained by analysing the harmonic component in the signal. The delay of four wave plates is measured, that of three wave plates among them has been precisly measured by electrooptic modulation method. By comparing the results of the two ways, it is found that they have little difference. The oscillation phenomenon in the delay of the isinglass wave plate is found,which had been proved in the reference 3. The experimental results show the phase-modulated ellipsometry is of significance.

Spectral analysis on the transmission fluctuations is a new method for measurements on large particles. Radiated by a narrow beam, the flowing particle system at a uniform velocity causes the transmission signal fluctuate violently. Both the average transmission and the spectrum of transmission fluctuations can be obtained statistically, which include the complete information on particle size distribution and particle concentration (volume fraction). By using the inverse algorithm, the particle size distribution may be extracted from the spectrum of transmission fluctuations and consequently the particle concentration can be achieved from the average transmission. The theoretical expression of the transmission fluctuation spectrum is established, for monolayer of mono-dispersed particles. The theory is then extended to a 3-dimensional system of mono-dispersed particles and further to that of poly-dispersed particles. Experimental test and simulation on dilute particle systems with particle diameters in the range 32～425 μm are performed, which shows that the method is valid to measure the particle size distribution and particle concentration simultaneously.

A novel measurement method for the torsional vibration of high-speed rotary machine is proposed, based on laser Doppler technique and the principle of optical heterodyne. The article analyzes its principle, worked out the mathematical model of the optical configuration, and tests the feasibility and accuracy respectively. The laser beams launch at two points of the same section of the rotating shaft. The reflected beams are shifted at frequency domain according to Doppler effect. The frequency shift is proportional to the rotating velocity. With the proper optical configuration, the beams of two successive times are heterodyned on the photo-detector, and the frequency of the photoelectric current is proportional to difference of velocity at the two times. By making the interval of the two successive times small enough so that the angular acceleration is measured directly. Compared with other measurement methods based on laser Doppler technique, the angular acceleration is measured directly, therefore the measurement of torsional vibration is more real-time and dynamic range is extended greatly. It is essential to the monitoring running state and failure diagnosis of large-scale gyration machine with high-rotational velocity.

In order to get higher accurate phase of the double spacial frequency fringes, a method that employs the orders of projected fringes as axises to establish fringe-order-coordinate is presented. It makes the description of the fringe orders (phase) and their errors very direct and simple. In the fringe-order-coordinate, Bayes estimation method is used to amend the phase, and the more accurate phase is obtained. Emulational tests and actual experiments show that this method is effectual. With the application of this amending method in the actual experiment, the MSE of phase decrease from 0.014 rad to 0.009 rad, and the mean square error (MSE) of height decreases from 0.058 mm to 0.041 mm.

Overlay is one of key performances for modern lithographic projection systems with high-precision, a novel method for in-situ measuring the overlay performance of a lithographic system with the marks is presented. In the method, many mirror-symmetry focus calibration using alignment procedures (FOCAL) marks are printed on a wafer. The printed images of these marks are aligned by an optical alignment system and the alignment positions of the images are recorded. From the positions, overlay performance can be obtained. Through the experiments, it is demonstrated that the overlay errors, intrafield parameters and interfield parameters can be obtained with high accuracy. Compared to the XY-SETUP technique, the method avoids the dependency of measurement accuracy on the corrections of axial aberrations and makes it much simpler to perform a full-scale evaluation of a lithographic system.

A high performance algorithm to optimize the wafer exposure field layout is presented to improve the lithographic machine performance. The size of the best exposure field is calculated from the size of the chip, extremely close to the maximum size of the exposure field of the lithographic machine, so the utilization ratio of exposure system is increased. The offset layout of the exposure field is implemented to decrease the overlap of exposure fields in edge area of the wafer, which increases the lithography throughput. With the adoption of two optimization criterions, throughput prior and yield prior, the exposure field layout is optimized for both throughput and yield. And the algorithm is used in the exposure process in a wafer fabrication laboratory, with the throughput prior criterion, and the practical viability of this approach is validated while taking the parameters of the chip production as the example. The overlap of the exposure field decreases by 10%, while the number of inner fields keeps steady. For the yield prior criterion the overlap of inner field increases by 10%, while the total number of exposure fields decrease accordingly and the chip's throughput and yield are enhanced.

By analyzing the light intensity of the ring laser gyro (RLG) and its performance changes with variations of the environmental temperature, a new method is put forward. The angle controllers of a novel cavity length and angle control mirror-light path and optical path length control mirror are used to realize the closed\|loop control of the ring laser of RLG in a similar way to the dither frequency stabilization (hill climbing). The modulation signal conflict is settled by controlling all the functional units in time-sharing control means. The optical path length and the light path of the ring laser are stabilized when the environmental temperature changes. The experimental results show that in the whole range of -40 ℃~+60 ℃, in the condition that the original intensity fluctuation reaches 30% or the intensity is very feeble, the light intensity variation is modulated within ±3% with the controlling system, and even stabilized light intensity level is kept when the original intensity is zero under extremely high and low temperatures. The system resolves the environmental temperature influences on the RLG.

With uniformly side-around arranged compact pumping structure, high conversion efficiency operating of diode-laser-array side-pumped Nd∶YAG ceramic laser was demonstrated. Using a 5 mm×75 mm sized Nd∶YAG ceramic rod with 1% dopant fraction of number of atoms, the laser generated a quasi-continuous-wave (QCW) output with an average power of about 230 W at 1064 nm, and the output coupler transmission was 22%, which corresponded to the calculated result of 22.2%. The optical-to-optical conversion efficiency was 52.4%, and the slope efficiency was 61%. Besides, the pulse width of the beam was detected to be 160 μs, the line width was less than 0.8 nm， and the beam divergence angle was 16 mrad. The experiment results showed that the output power Nd∶YAG ceramic laser was comparable to that of Nd∶YAG single crystal laser.

Single-scattering diffuse backscattering Mueller matrices expressions based on Stokes vectors of the backscattered light are derived when a narrow pencillike beam is perpendicular to the surface of the plane of turbid medium. Using Mie theory of light scattering, the characteristic distribution pattern of single-scattering Mueller matrices, and the correlation with particle number density and particle diameter are discussed. The results indicate that the azimuthal variations gradually disappear when the particle number density increases, and m22, m33, m23, m32 elements are the more pronounced azimuthal variation characteristic. For azimuthal angle =45°, m22 and m33 decrease with increasing particle size if particle size parameter is smaller than the characteristic parameter, otherwise, m22 and m33fluctuate and gradually increase. And when the particle size and particle number density are varied, all of the Mueller matrices elements intensity patterns' dependence on the radial are the same, exponential, decaying, for any azimuth angle.

The interaction between the specimen and the scanning probe in the near-field interferometry was modeled with a coupled point-dipole system, and a set of equations of different electric field components were used to calculate the intensity and phase images of the specimen. It was found that, the resolution of the near-field images was strongly dependent on both the polarization direction of the reference light and the position of the far-field detector and vice versa. For a specimen under given illumination, its near-field images changed remarkably with the slight adjustment in the detector position or the polarization of the reference light. In the case of evanescent illumination, the resolution of differently polarized images increased with the view angle of the detector to the specimen, and highly accurate images could be realized only when the detector was placed at a large enough view angle. Among the images of differently polarized components, only the image polarized normally to the specimen surface indicated the specimen in details with enough intensity.

Two-photon three-dimensional optical data storage is an effective method for high-capacity data storage. The beam of light passes through two different media (such as air and storage material)in writing three-dimensional information and influences the optical aberration and storage effect. So it is important for analysing the effect of each optical parameter of system on the optical aberration and storage impression of refractive index mismatch in the theory and experiment. First, the model of the optical storage system is set up， in the condition of the parallel flat and via wave aberration function deducing， the five primary Seidel aberrations: spherical aberration， coma， astigmatism， field curvature， and distortion， are gotten, and simulations and analyses on the primary aberration in the two case of horizontal and tilted storage material are performed. Simulation and experiment proved that the slope of optical aberration increased with the increase of storage depth; if numerical aperture got bigger optical aberration augmented quicklier.

Three types of opto-electronic hybrid neural networks are introduced. The errors induced by diffraction in these networks are emphasized and analyzed with computer simulation using the experimental parameters and input data， where diffraction, optical information processing and neural network theories are applied. It shows near-field diffraction induces large magnitudes of relative errors. When near-field diffraction and far-field diffraction are employed together, the errors are different according to the display modes. When the display mode is a small image, the magnitudes of relative errors are very large. However, when the display mode is a complicated image, the magnitudes of relative errors are small. The feasibility of using linear regression to calibrate the output data is discussed. It is found that linear regression can reduce the errors for about one magnitude. According to the analyses, the errors induced by diffraction can be minimized to a low level in the experiments, and thereby, the recognition rates can be maintained at a high level (larger than 97.7%).

Binary joint-transform correlator (JTC) is used to implement optical parallel Dempster's rule of combination computation. The corresponding theory and method about how to implement the orthogonal sum of the basic belief assignment has been discussed. The corresponding theory has been discussed, and computer-simulation results are presented, which prove this method is practical. Compared with the method John Caulfield proposed implementing an outer product by two crossed 1-D acoustooptic cells, this one benefits from the advantages of high accuracy because of the digital representation of the multiplied numbers and of the simplification in the optical process of Dempster's rule of combination computation because of the direct result of the orthogonal sum of basic belief assignment by the appropriate arrangement of the binary spaial code in the input plane of the joint transform correlator.

The digital speckle fringe pattern is modulated by spatial carrier in large-shearing electronic speckle pattern interferometry (ESPI). The carrier is introduced by rotating the reference plane a small angle. When Fourier transform is used to demodulate the modulated fringe pattern, the phase of deformation is calculated and furthermore the deformation is measured accurately. The principle of carrier modulation in large-shearing ESPI is discussed. The theoretical analysis shows that the frequency of carrier used for modulation is proportional to the rotational angles of reference plane. Therefore, the fringe patterns with different displacements can be modulated easily by rotating reference plane different angles. A typical experiment using a centrally loaded clamped circular plate is completed. The experimental results prove that the carrier introduced by rotating the reference plane in large-shearing ESPI can modulate speckle pattern very well and the displacement fields can be obtained effectively. The optical modulation system has some advantages: the system is simple and has no special reference beam for interference. And the carrier pattern has higher fringe visibility. This technique presented here would extend the application range of ESPI and have some applications in practices.

With a confocal Fabry-Pérot (CFP) cavity as a transfer cavity, the single-frequency laser can be stabilized to atomic or molecular lines when no atomic or molecular reference lines are available. In the experiment a CFP cavity as the transfer cavity is locked to homemade 852 nm grating-external-cavity diode laser whose frequency is stabilized to 6S1/2 Fg=4-6P3/2 Fe=5 transition of cesium atom via RF frequency-modulation spectrum and saturated absorption spectrum. The frequencies of external-cavity diode lasers at 908 nm and 830 nm are locked to the stabilized transfer cavity via Pound-Drever-Hall RF sideband technique, and thus stabilized with respect to the transition of cesium atom. Estimated from the locked error signals, the corresponding frequency jitters of diode lasers at 852 nm, 830 nm and 908 nm are less than ±540 kHz, ±340 kHz and ±60 kHz in 20 s, respectively; while the frequency fluctuation of the CFP cavity to the 852 nm laser is in ±550 kHz.

The damage and ultrafast dynamics processes induced by 800 nm femtosecond laser in ZrO2/SiO2 800 nm 45° high-reflection coating are studied. The damage morphology, and threshold fluence as a function of pulse duration and the dependence of ablation depths on the pulse fluences are measured. The damage threshold increases from 0.35 J/cm2 to 1.78 J/cm2 with pulse duration enlarged from 50 fs to 900 fs. The ablation depth is linearly proportional to the logarithm of pulse fluence, which suddenly reaches hundreds of nanometers once the laser fluence is above the threshold, and the layered ablation shows. By using of a pump and probe experimental system, the time-resolved reflectivity irradiated by high intensity laser pulses in the reflector ZrO2/SiO2 is measured. The results indicate that the optical properties of the surface layer of high-reflection coating play a very important role during the damage process.

Monocrystalline ZnO films with mosaic structure are successfully grown on Si(111) substrate by atmospheric-pressure metal-organic chemical vapor deposition (MOCVD). Low-temperature AlN buffer layer is introduced between the films and substrate to protect the substrate from being oxidized and eliminate the thermal and crystals lattice mismatch. Only appear diffraction peaks of Si(111) plane, ZnO(000l) index planes and AlN (000l) index planes [including the (0002) and (0004) planes] in double-crystal X-ray diffraction 2θ/ω scanning curve. The lattice constant of C direction is 0.5159 μm, which indicates ZnO film is suffering from the tensile stress. The full wide of half maximum (FWHM) of ZnO (0002) and ZnO (1012) double-crystal X-ray diffraction ω-scan peaks are 460″ and 1105″ respectively. The crack density of ZnO surface is only 20 cm-1 by optical microscope graph determination. Mean square root roughness is 1.5 nm from the 3 μm×3 μm atomic force microscope scanning graph. The in-situ laser reflectance trace shows that a quasi-two-dimensional growth mode is obtained and the growth rate is 4.3 μm/h. Free exciton emission and binding exciton emission accompanied by their longitudinal optical phonon replicas are observed from the photoluminescence (PL) spectrum at 10 K. All the results show that AlN buffer layer is an effective route to obtain high quality ZnO film on Si (111) substrate by MOCVD.

To study the damage mechanism and micromachining of 6H SiC, the femtosecond (fs) laser pulse is focused on the front of the sample by a 10× microscope objective lens. The ablation pits are measured with scanning electron microscope (SEM) and optical microscope (OM). The experimental results indicate that high quality micro-structures can be fabricated in SiC crystal. Based on the dependence of ablation area on laser pulse energy, the threshold fluence is determined accurately, and the wavelength dependence of SiC ablation threshold fluence is given. The threshold fluence of SiC is found to increase with the incident laser wavelength from 0.29 J/cm2 to 0.67 J/cm2 in the visible region, while it remains almost constant (about 0.70 J/cm2) for NIR laser. For the NIR laser pulses, both photoionization and impact ionization play important role in electronic excitation, while for viaible lasers, photoionization is dominant.

The propagation of linearly and circularly polarized high-intensity femtosecond laser pulses in bulk media is experimentally investigated. The spectrum broadening is different after propagation in bulk media for different input pulse polarization. The pulse width is compressed after the dispersion compensation and the different compression results when using linearly and circularly polarized laser pulse are also compared. The effect of the input laser pulse energy on the compressed pulse width and its envelope is invertigated. The shortest compressed pulse width is 21 fs by using linearly polarized input pulses, and it is 22 fs for circularly polarized laser pulses. The experimental result shows that the spectrum broadening and dispersion compensation method is also valid when compressing circularly polarized laser pulses, which will benefit higher-energy femtosecond laser pulse compression. When the dispersion compensation is the same, the relationship between the compression result and the input laser energy agrees with the theoretical analysis.

Based on the Franz-Nodvik mode, a detailed numerical analysis of the contributions of the fluence of the pumping light and signal light as well as the crystal parameter to the output pulse energy and conversion efficiency is presented. Comparison between the theoretical result and the experimental result shows that the theory is quite valid. The calculation indicates that in the system of chirped pulse amplification, there is an optimum fluence of pumping light. Reasonably choosing the fluence of pump light will not only advance the conversion efficiency effectively, but also remarkably reduce the demand of signal light energy and the output instability caused by the fluctuation of signal light energy.