A new compression scheme of interferential multispectral images is proposed based on the analyses of large aperture static imaging spectrometry. There are two novelties in the scheme

Satellite multi-spectral image series are produced by image location, so there is strong correlation between neighbor images. The difference from usual image series is that there are apparent shift features between the image series. In order to use this feature to improve the compression efficiency and decrease the complexity of the coding system, a novel multi-spectrical image series coding algorithm is proposed. When the relative shift between two images is detected by the match algorithm in wavelet field, the error image between the images is coded in stead of the following image coding such that the whole coding efficiency is improved. The algorithm is simpler than the one based on three-dimensional wavelet transform, and the disadvantage with large memory requirement and big coding delay is overcome. The simulation results show that better effect is achieved than the one based on three-dimensional wavelet transform.

Based on the characteristics of numerical reconstruction for lensless Fourier transform hologram, a set of digital image processing algorithm is presented to eliminate zero diffraction component and “twin image”. This method is different from the former one and consists of three steps

Digital watermarks have been proposed in recent literature as a method for copyright protection of multimedia data. Digital watermark is invisible mark embedded in digital image, video or audio documents, by detecting the information of watermarking to protect copyright of multimedia data. Most published watermarking methods are based on two-dimensional (e.g., image or video) approaches. A new algorithm of digital watermarking in three-dimensional (3-D) space based on the methodology of virtualoptics is presented. The watermarking algorithm can offer an effective solution to the issue of copyright protection of multimedia data. Numerical experiments show that the embedded watermark is robust to signal manipulations, such as superposing noise, cropping and lossy compression. Because the concept of virtualoptics is introduced, the proposed scheme has good security. There are large degrees of freedom existed for designing a security lock(s)/key(s), including multiple lock(s)/key(s) and multi-dimensional lock(s)/key(s). Numerical simulation results show a good agreement with theoretical prediction.

By using Fresnel diffraction and holographic theory, some detailed analysis of the recording, numerical reconstructing method and the characters of the reconstructed image about lensless Fourier digital holography and phase-shifting digital holography is given. The experimental result shows

Now, higher loss in polymer modulators is one of important problems to be solved in the research of electrooptic modulators. The waveguide loss plays an important role in the insert loss. Especially, modulators will inevitably have higher waveguide loss seeking high EO coefficients. It is an important topic to improve the modulators' loss. An idea is proposed that light-wave modulated only in claddings. Analysis shows that this method can decrease largely waveguide loss, so it helps to improve the insert loss. Because the light power in claddings is weaker than one in core, the overlap will weaken when modulation is realized only in cladding. It is found that optimizing the dimensions of waveguide can largely reduce this weakening. Calculation makes the optimum overlap to 0.89. The length of interaction can increase with the decreasing of waveguide loss, which is helpful for decreasing half-wavelength voltage.

Starting from the nonlinear Schr??dinger equation, the evolution equations of the perturbation phase and magnitude in optical fibers have been deduced under small signal approximation. Using the obtained solutions of the perturbation phase and power gain, the influence of gain of perturbation propagation is investigated for different initial phases and frequencies. The results show that at the initial stage, the gain is greatly affected by the perturbation initial phase. Depending on the initial phase, all the unstable perturbations have the chance to claim a common maximum gain that is frequency independent. In the normal dispersion region, as well as in the abnormal region where the perturbation frequency is larger than the cut-off frequency and modulation instability is thought not to occur, the perturbation gains oscillate with the propagation distance. In the abnormal dispersion region when the perturbation frequency is smaller than the cut-off frequency, where modulation instability is thought to occur, the gains of perturbations with different initial phases while with the same frequency will evolve to the same positive value through different ways.

Optical phase conjugation (OPC) is a promising optical technology that can compensate for dispersion and nonlinearity effects simultaneously in pulse propagation through a single-mode fiber. In optical fiber communication system with midway OPC employing dispersion management, effects of higher-order dispersion are analyzed theoretically. Dynamic evolution of ultra-short Gaussian pulses in the system is simulated numerically. The results demonstrate that the higher-order dispersion-managed OPC system can not only compensate for all nth-order dispersion including even order and odd order, but also eliminate the interaction of pulses, hence improve the compensation performance for the distortion.

Gain characteristics of small signal in L-band Erbium-doped fiber amplifier (EDFA) are analyzed with the numerical simulation using Giles model. The results show that the optimal fiber length varies with input signal wavelength, in general, shorter wavelength demands shorter fiber length. 7 m and 9 m L-band EDF are employed to construct L-band EDFAs respectively. Evaluation of EDFAs is made by experimental comparision of their amplified spontaneous emission (ASE) spectr, gains and noise figures (NF), and the results show that the fiber length is crucial for proper gain characteristic. On the basis of these numerical, experimental study and also considering their C-band characteristics, optimal ASE spectrum related to the best behavior of L-band signal amplification is presented.

Polarization mode dispersion (PMD) causes severe signal impairments in high bit-rate optical transmission systems. It is important to know how does the degree of polarization (DOP) change with differential group delay (DGD) when DOP is feedback information for adaptive PMD compensation. The mathematical expression between DOP and DGD is derived for arbitrary waveshape and Gaussian pulse. Then the relationship of DOP-DGD with different power splitting ratio, width of pulse, and amount of pluses is theoretically analyzed. Experiments with 10 Gbit/s return-to-zero/non-return-to-zero (RZ/NRZ) pseudo random sequence are conducted and comparison between theoretical calculation and experimental results is made. The comparison showed that they matched closely. It is also verified that DOP was independent of chirp and fiber dispersion in experiment, which made the theoretical analysis of DOP-DGD easy.

In the continuous tuning process of the electric tunable liquid crystal Fabry-Perot filter (LCFPF), the phenomenon of the jumping of the transmission peak spectrums is found. And this phenomenon influences the monotone tuning performance of the device. Analysis for this phenomenon shows that it's due to the change of the interference order of Fabry-Perot cavity in the tuning process. And the relationship between the transmission peak wavelength λ, the interference order m, and the index n of the medium in the Fabry-Perot cavity is given, as well as the effective tuning range of the device. When the length of the cavity is 10 μm, the tuning range will cover the whole C Band with the index difference of only 0.05. The experimental result is in agreement with the numeric simulation. At last, some approaches are presented in order to improve the monotone tuning performance of the device.

Optical and thermal properties of a novel optically readable thermal imaging focal plane array (FPA) are analyzed, including reflectivity of two reflective surfaces of Fabry-Perot micro-cavity, temperature response of movable micromirror, noise-equivalent temperature difference (NETD) and thermal response time. Theoretical analysis indicates that when the two reflective surfaces of Fabry-Perot micro-cavity have same reflectivity of 0.5 better reflective light intensity and modulation degree can be achieved and that reflective light intensity varies linearly with the displacement of micro-mirror in the range of 0～λ/4. Temperature response of the movable micromirror, NETD and thermal response time are estimated as 2.05×10-4, 19.5 mK and 1.54×10-3 s, respectively.

The theoretical model of the composite microcantilever used in the IR radiation is given based on the theoretical analyses, the infrared absorption characteristics of carbon nanometer tube (CNT) is studied by experiment. A tri-layer microcantilever sensors with CNT film is designed to detect the infrared radiation. The detector is based on the characteristics that the resonance frequency of composite microcantilever shift with the temperature. So, the resonance frequency of microcantilever can be used as sensitive parameter to measure the infrared thermal radiation. The theoretical analysis and simulation results show that this new model can attain power sensitivity of fWtwo orders of magnitude better than the sensitivity of traditional statical ones. Furthermore, the micro-electro-mechanical system (MEMS) sensor could be compatible with integrated circuit technology.

The propagation characteristics of the photonic crystal fibers (PCF) are numerically analyzed using the finite-difference time-domain (FDTD) method. The dispersion characteristics and the birefringent properties of PCF with any given transversal structure can be obtained by the FDTD method. To improve numerical precision, the anisotropic perfectly matched layer (APML) is used as an absorbing boundary condition in computing. The propagation characteristics of PCF are wholly determined by its transversal structure. One type of PCF with symmetric air-silica structure and two types of PCF with asymmetric air-silica structure are analyzed numerically. Numerical results show that it may exhibit high birefringence in a properly designed asymmetric PCF (whose birefringence can be as high as 0.07). It is shown that the FDTD method can be effectively used in the analysis and design of the PCF with special dispersion and polarization characteristics.

The light wave propagation in one-dimensional photonic crystal with multiple defects is studied by eigen matrix method. There are defect modes in the bandgaps. The defect modes are greatly related with the generating form, optical thickness and refractive index of defect. With the optical thickness of impurity changing, the position and number of defect modes change. Keeping the optical thickness of impurity constant, the changing of refractive index of defect leads to the changing of intensity of defect mode, and there is a maximum among them. If the position of impurity arrange in equal distance, then the defect modes arrange in equal distance, too. The creation of defect modes can broaden the bandgaps, especially when refractive index of defect is much different from that of periodical medium. The defect of air can make transmission rate up to one.

This intensity distribution of rainbow at different temperature and its application in grain size measure are investigated theoretically and experimentally. The results show that scattering intensity distribution of droplet is not simply second-order rainbow distribution while refractive index is around 1.33 with scattering angle between 117° and 134°, but the interference distribution between second-order and fifth-order rainbow. According to the spectral characteristics of scattering intensity, the method of reconstructing intensity of the second order and fifth order rainbows from interference intensity spectrum is presented

Stroboscopic shearing speckle interferometrya method for investigation of vibrations is proposed. While the shearing lens is used, the speckle shearing picture is performed, to improve the precision of measuring. The fundamental principle of stroboscopic shearing speckle interferometry is discussed. For a specklegram of stroboscopic shearing in whole field analysis, the analytic formula of the distribution of the mean intensity of light and experimental result are given. Both theory and experiments show that stroboscopic shearing speckle interferometry is obviously superior to time-average method. In the whole field filtering, the information of vibration at any moment can be easily obtained ,and images obtained are quite clear.

Starting from the ultrashort pulsed Gaussian beams, a family of solutions are found theory in by using the properties of Hermite function, which represents a new family of ultrashort pulsed beams called ultrashort pulsed elegant Hermite-Gaussian beams. These pulsed beams with a certain frequency have a nearly elegant Hermite-Gaussian profile, an arbitrary temporal shape, and a common diffractive distance. These solutions can be even used to describe the ultrashort pulsed beam whose pulse duration is shorter than one optical oscillation period. The ultrashort pulsed elegant Hermite-Gaussian beams and their propagation properties in free space have been studied in detailedly, such as intensity on the axis, polarity reversal, pulse time delay etc. The spatial singularity of the pulsed beams is also discussed when the slowly varying envelope approximation is introduced.

The Jaynes-Cummings models with a time-dependent atom-field coupling under the initial squeezed state of the optical field are discussed. Especially, the evolution of the atomic population inversion, and the squeezing properties of the field with time are studied. Further, the effects of the atom-field coupling coefficient, the squeezing parameter and phase on these properties are also discussed, and some results are obtained. With the increasing of squeezing coefficient, the frequency of collapse-revival in atomic population inversion are enhanced, and the amplitude of revival decreases, in the same time, there are no squeezing. When atom-field coupling coefficient changes rapidly, it will fasten the collapse-revival. With the increasing of squeezing phase, it will delay the collapse-revival and decrease the amplitude of revivals.

The atomic parameters of the transmition 2s22p6-2s22p53s of Ne-like ions with nuclear charge Z from 13 to 27, including energy level, oscillator strength, spontaneous radiation deay time, level life, electron-collision excitation cross-section, electron-collision excitation rate and so on are obtained by using Hartree-Fock-Roothan (HFR equations). The spectral line enlargement and population inversion formation in Ne-like Ar at 46.9 nm laser transition and the probability of gain generation of other spectral line are discussed with the help of the atomic data. The dependence of oscillator strength on nuclear charge is illustrated. The discharge parameters for Ne-like Ar and Ne-like Kr X-ray laser with capillary discharge are also given based on the obtained atomic parameters. The range of the initial pressure of Ne-like Ar gas is 30～90 Pa and the peak value of the discharge current is 10～50 kA for lasing of Ne-like ions.

A method for flatly ultra-broadened supercontinuum generated in anomalous dispersion region of a dispersion-shifted fiber is introduced. Supercontinuum generated in anomalous dispersion region of dispersion-shifted fibers due to compression effects of higher-order soliton is studied in detail through numerical simulations. The results show that the dispersion slope of dispersion-shifted fiber is crucial to supercontinuum generation. It also shows that both the peak power and pulse width of pump pulse affect the spectral width and flatness of supercontinuum markedly, but higher-order nonlinear effects can be ignored. A flatly ultra-broadened supercontinuum near 300 nm with the spectral-intensity fluctuation less than ±2 dB is obtained by optimizing the relative parameters.

The characteristics of the switching curves of Gaussian-Constant-Gaussian varying coupling coefficient nonlinear directional coupler (GCG NLDC) are analyzed in detail. The influences of the parameters of the coupler's subsections on the switching curves are studied. It is found by numerical simulation that the existence of sidelobes of the switching curves and the value of the switching powers are related to the parameters of all the subsections. The magnitude of sidelobes of the switching curves is decided by the relative length of bent coupler L/b and the quotient of conventional NLDC, the sidelobes of the switching curves have close relations with the oscillation of the Fourier transforms of the coupling functions.

Cutoff wavelength is an important parameter for optical fiber. The cutoff properties of hole-assisted optical fibers that comprise six air holes in cladding are investigated by a full vector orthogonal function model. It is shown that when the material dispersion is neglected, the effective cladding index is a function of wavelength due to the introduction of air hole into the cladding of fiber; the effective cladding index is determined by the relative air hole size in cladding, and independent of the relative index difference of doped core. The fundamental mode cutoff occurs only if d/Λ is large enough, which is similar to the case of traditional W-shaped refractive index fibers. The cutoff wavelength of each mode decreases as the relative air hole size d/Λ increases from 0.2 to 0.9; the slope of cutoff wavelength versus d/Λ curve decreases as the order of mode increases.

It is very difficult to multiplex fiber Fabry-Perot sensor because of its fundamental limitation. When serially multiplexed, it is especially complicated due to mutual affection between sensors. To address this problem, the simplest case of two fiber Fabry-Perot (F-P) sensors multiplexed in series was discussed. According to the principle of double beam interference, the output of serially multiplexed sensors was derived, and its characteristics was discussed in detail. Fourier transform was adopted to transform the output from wavelength domain into spatial frequency domain. The dissatisfaction of conditions for using Fourier transform signal processing of fiber Fabry-Perot senor was discussed, and distortion of Fourier transform was analyzed. With the simulation of Fourier transform, the eigenvalues of two fiber Fabry-Perot sensor were demultiplexed in spatial frequency domain. In addition, experiment of serial multiplexing two fiber Fabry-Perot sensors was carried out. The experimental results showed that, Fourier transformation in demultiplexing serially multiplexed fiber Fabry-Perot sensors was valid, and mutual affection between two fiber Fabry-Perot sensors was less than 5%.

To demodulate wavelength shift of fiber Bragg grating, a new direct-demodulation technique is demonstrated. An unbalanced fiber Mach-Zehnder interferometer comprising a 2×2 coupler and a 3×3 coupler is used as wavelength discriminator for fiber grating, 120° phase difference between the three outputs of interferometer. The wavelength shift of fiber grating can be demodulated directly by calculating three outputs of interferometer. Demodulation is fulfilled by software when signals are sampled and inputted into computer. The minimum detectable dynamic strain of 0.51 nε/Hz1/2 can be achieved when the unbalanced length of the beams is 5 mm. The spectra of the demodulated signal and the relationship between the input and output are also given.

The temperature response and sensitivity enhancing technology concerning fiber Bragg grating (FBG) is analyzed. A special kind of heat-resistant polymer is used to coat FBG to enhance the temperature sensitivity of FBG sensor. The polymer coating and solidifying process of FBG is improved. A kind of heat-conducted adhesive is used to reduce the friction between tube and polymer, and eliminate the chirp effect of FBG caused by the asymmetric shrinkage of polymer during the package process. The experimental results show that the FBG sensor can measure temperature from 20 ℃ to 180 ℃. The temperature response sensitivity is 0.05 nm/℃ from 20 ℃ to 130 ℃ and 0.22 nm/℃ from 130 ℃ to 180 ℃, respectively, and the response trend line has good linearity in both areas. This kind of sensor is easy to fabricate and can be used for high temperature measurement.

A simple dynamic chemical etching device based on siphon principle is developed for fabrication of optical fiber probes which are commonly used in near-field optical microscopy. Probes with various shapes have been fabricated because the taper length and taper angle of the probe could be efficiently controlled through adjusting the speed and direction of water flux while etching. Compared with traditional static chemical etching, this method has the advantages such as reproducibility, controllability, convenience, lower cost, and it can make the probe surface smooth. Probes with apex diameter from 50 nm to 300 nm and taper angle from 16° to 65° have been achieved by this method. Moreover, double-taper probes with apex diameter less than 50 nm and taper anger larger than 125° could also be obtained by selecting appropriate water level difference. The mechanism of the chemical etching is discussed.

Analytic description of disc diffraction model is obtained using elliptical pits model and Babinet principle. The influence of disc radial tilt on tracking servo system is analyzed and computed focusing on three common disc types. Tracking error offset caused by radial tilt is on DVD-ROM using DPD (Differential phase detection) method, 0.15 μm/° using push-pull method. Tracking error signal with DPD method exhibit analyzed better performance than push-pull method against disc radial tilt under the same disc parameters. Finally relationship between groove depth and tracking error signal offset of DVD-RAM disc is analyzed.

Titanium thin films have been prepared on borosilicate glass (BK7) substrates by ion beam sputtering technique with different voltage of sputtering, and the effect of energy process on the surface morphology is studied with the help of atomic force microscopy (AFM). Numerical correlated calculations show that the surface morphology appears to be fractal under these deposition conditions. Based on the fitting of height-height correlation functions of thin films to the phenomenological formula of fractal surface, the parameters such as fractal dimension, lateral correlation length and interface width are all obtained. The results show that the interface width increases but the fractal dimension decreases with the energy of incident Ar ions when the voltage of the sputtering is between 300 V and 700 V. Furthermore, the growth mechanism is studied based on the obtained fractal dimension under the different voltage of sputtering.

In order to improve the technique of making microstructure on azobenzene polymer film, a new experimental method is presented, based on photoinduced isomerization and photoinduced birefringence of an amorphous copolymer containing azobenzene groups. Two-dimensional square microstructure is written on the azobenzene functional polymers film by two laser beams in a step-by-step procedure, so the pattern can be controlled in a convenient way. A new theory of photoinduced isomerization and re-orientation field is put forward. When azobenzene polymer film is irradiated by two beams and the polarization of these two beams are perpendicular, photoinduced isomerization and re-orientation field will occur and the interference can also take place. And the interference induces the formation of the grating on the surface of azobenzene polymer film. The results of the experiment and computer simulation are compared.

Undoped and indium-doped zinc oxide films are deposited on Si substrates by radio frequency reactive sputtering technology. Glancing X-ray diffractometer (XRD) measurement indicated that In-doped sample is ZnO films. The structure, surfaces morphology and photoluminescent spectra of the sample are characterized by X-ray diffractometer, atomic force microscopy and fluorescent spectrophotometer, respectively. The effect of In-doping on the structure and photoluminescent properties of the films is analyzed. Compared with undoped ZnO film, In-doped ZnO film has highly c-axis oriented and the small lattice mismatch (0.16%). Surface of doped thin film is smooth and flat; the maximum roughness surface of sample is only 7nm. The blue-violet emission bi-peak locating at 415 nm and at 433 nm is observed in photoluminescence spectra of indium-doped ZnO films at room temperature. The mechanism of blue-violet emission bi-peak was discussed, and the blue-violet emission bi-peak is assigned to come from the In substitute impurity and Zn interstitial defects of ZnO.

A new type of AgInSbTe phase change thin film is prepared by direct magnetron sputtering. The temperature for crystallization obtained by differential scanning calorimetry (DSC) is 193.92 ℃. The undeposited AgInSbTe is amorphous by the measurement of X-ray diffraction (XRD), while the phase change film is heated at 200 ℃, the peak could be observed by XRD, which indicated that the amorphous state had become a crystalline state. The variation of the absorption, transmission and reflectivity of the amorphous and crystalline phase change film with the wavelength are studied. The recording performance is measured at the 650 nm laser, the influences of the recording power and recording pulse on the reflectivity contrast are analysised. Under the same pulse, the higher the writing power is, the higher the reflectivity contrast is, and under the same writing power, the reflectivity contrast increased with pulse during increased. The results indicate the new type AgInSbTe phase change film irradiated by laser had a high reflectivity contrast, and a good recording performance could be obtained.