Mobile differential optical absorption spectroscopy (DOAS) is used to study regional distribution of SO2 and NO2 column density in the North China Plain and its impact on the Beijing region under different wind fields during June and July in 2013. High SO2 and NO2 concentration values around Shijiazhuang, Baoding and Jinan areas are observed simultaneously, indicating that there are industrial emission sources in these areas. Average values of SO2 and NO2 around the Shijiazhuang area on June 11th are 1.29×1017 molecule·cm-2 and 3.59× 1016 molecule · cm- 2, respectively, which are 3.8 times and 3.6 times of the average in low concentration area respectively. Shijiazhuang- Baoding- Beijing direction can be a transportation pathway in steady south- west wind field, and the conclusion is verified by the Hysplit model. NO2 column density obtained by mobile DOAS is compared with that by ozone monitoring instrument(OMI), and the good agreement is found. The satellite observation also proves existence of the transportation pathway in south wind field. The experimental results show that mobile DOAS plays an important role in detecting emission sources, rapidly obtaining regional distribution of air pollutants and validating satellite data.

The geostationary ocean color imager (GOCI) can provide high temporal resolution (1 h) and spatial resolution (500 m × 500 m) observations. Validation of the GOCI- derived ocean color products is a key for the application of the GOCI data. Based on matched-up in-situ measurements in Liaodong Bay during spring, the GOCIderived remote sensing reflectance is retrieved by four atmospheric correction algorithms，including Korea Ocean Satellite Center(KOSC)correction, management unit mathematics models (MUMM) correction，ultraviolet correction (UV-correct) and near infrared correction (NIR-correct), and the GOCI standard chlorophyll-a concentration and suspended particulate matter concentration are validated. In addition, the stability of the GOCI-derived ocean color remote sensing products under different observing time is evaluated. The results show that the remote sensing reflectance retrieved by KOSC algorithm has the best agreement with the values of in-situ observations, with mean absolute percentage deviations (APDs) of 27.16% (412 nm), 16.03% (443 nm), 13.73% (490 nm), 15.99% (555 nm), 12.83% (660 nm), 12.35% (680 nm), 27.68% (745 nm), 42.81% (865 nm) . The APDs of chlorophyll-a and suspended particulate matter products are respectively 29.75% and 26.40%. Moreover, compared with the ocean color products obtained from 10:30 to 12:30 (Beijing time), the consistency of GOCI-derived ocean color data obtained between in situ and in other observing time is much poorer.

The optical properties of the semi-closed Bohai Sea are complex due to significant terrigenous influence.The high temporal resolution data from geostationary ocean color imager (GOCI) makes it possible to explore thetemporal and spatial variability characteristics in the Bohai Sea, especially on the hourly temporal scale. A remotesensing algorithm is developed to retrieve the total absorption coefficients at 412, 443, 555 nm in the Bohai Sea basedon the GOCI band ratio of remote sensing reflectance. The algorithm is validated with independent field data. Thecorrelation coefficient between the retrieved and measured data is higher than 0.8, the average absolute percentagedifference (APD) is less than 13%, and the median absolute percentage difference is less than 9%. The performanceis better than that of previous typical algorithms (improvement over 10%). Based on the independent field-satellitematch-up data, the APD between the retrieved and measured data is less than 35%. On the basis of the GOCI image (May 4th，2015) and the new model, the spatial and hourly variation characteristics in the Bohai Sea are discussed.The results show that diurnal variability magnitude is mainly 0.2~0.3 m-1, and the maximum can reach 0.8 m-1. Highdiurnal variability values are found mainly in the Liaodong Bay, the Bohai Bay and the Laizhou Bay. The totalabsorption coefficient is analyzed together with tide data, and the variations are generally consistent with thecorresponding tidal height variations in the Liaodong Bay and the north of the Bohai Bay, though the minimal totalabsorption coefficient appears about 0~1.5 h later than the lowest tide moment. The total absorption coefficientshows significant negative correlation with tide height in the west of the Laizhou Bay. The total absorption coefficientand tidal height have no obvious correlation in the south of the Bohai Bay and the east of the Laizhou Bay.

The principle of differential column image motion light detection and ranging for acquiring atmospheric turbulence profile is described. Aiming at the large retrieval error in high-altitude turbulence of current Levenberg- Marquardt inversion algorithm, a novel inversion model with inequality path constrained is developed and the penalty function method is used to handle this model, thereby an unphysical solution by adding the information of highaltitude turbulence is avoided. Furthermore, in order to weaken the current algorithm dependence on initial value and priori knowledge, a new optimization strategy based on genetic algorithm is presented to locate initial value of current algorithm in global variable space. Typical atmosphere turbulence profiles are simulated with the modified algorithm and the current algorithm. The measured lidar data in Hefei is also analyzed. Results show that the modified algorithm can enhance the global search capability of iteration process and perform strong robustness against measurement noise, improving the retrieval precision and accurate quantification of high-altitude turbulence effectively. Moreover, the modified algorithm accelerates the convergence.

Underwater range-gated imaging is an important method of underwater photoelectric detection. In order to obtain sufficient illuminating areas, an efficient model of the irradiance spatial distribution is derived for Gaussian laser illumination in the water based on small angle scattering theory. Irradiance distribution measurements of a 532 nm laser source with different divergence are presented. The errors of the model become large when the divergence of the laser increases and the illuminating range extends, but the resulting predictions of the analytical model are in good agreement with the experimental measurement over wide range of illuminating distance by limiting the field of illumination to the circle region with a diameter of 1 m. These predictions can give enough precision for underwater range-gated imaging systems.

Infrared detector module is the key device of target detection and imaging system and its spatial resolution directly affects the imaging quality of detection system. When evaluating spatial resolution of detector modules, researchers usually adopt modulation transfer function (MTF), which is mainly influenced by the optical crosstalk of detector modules. The basic principle and configuration of infrared micron-spot test system, whose optical dispersion diameter is 30 μm are introduced. The line spread function(LSF)of detectors with different structures is measured by the scanning slit technique, and a corresponding model is built for analysis. Theoretical analysis and numerical simulation agree well with the experimental results, demonstrating that the profile of overlap electric region is photosensitive, which is the main factor to cause widened LSF, asymmetry, secondary peak and so on. The result provides a reference for the design of infrared detector modules to suppress light stray.

In order to solve the problems of speed and accuracy in computing the transmission of light field by the current two kinds of arithmetics, a novel arithmetic of matrix multiplication is proposed. The implementation idea and procedures are given. A simulation example is based on six Gaussian beams coherent combination. The result shows that the speed of fast Fourier transform is perfect, but the calculation is not very accurate. However, the integral arithmetic and the matrix multiplication arithmetic can all get exact far-field intensity distributions. The time costed by these two arithmetics is evenly matched. The former needs 15.7 h and the latter only needs about 2 s. The comparison proves that the matrix multiplication arithmetic has advantages of fast and accuracy, which improves the calculation efficiency considerably.

Spatial electronic scanned low- coherence interference long distance optical fiber sensing and demodulation system is investigated based the superluminescent light emitting diode (SLED). The compact beam path of demodulation interferometry is built by negative lens. The influence of the double SLED light sources power ratio on low-coherence interference fringe is analyzed and the power ratio compensation for transmission loss is discussed. According to the characteristics of double SLED light sources interference fringe, a double-threshold demodulation algorithm is proposed and a pressure demodulation experiment is conducted. Experimental results show that, in the pressure range from 50 kPa to 180 kPa, the demodulation precision of 0.24% to 0.40% F.S. is realized, where F.S. is full scale, and that the longest distance can reach 1.76 km. The results are in good agreement with the theoretical analysis.

A PbSe quantum-dot-doped fiber amplifier (QDFA) is realized. Using PbSe quantum dots as the gain media doped in the fiber, the broadband signal light ranging from 1250 nm to 1370 nm is amplified in an all-optical framework，which consists of the quantum- dot- doped fiber, 980 nm pumping laser, wavelength division multiplexing, and isolator. For the diameter of 50 mm quantum-dot-doped fiber, there is evidence to show that the excitation threshold is 62 mW, the bandwidth is 120 nm at -3 dB, the flattened bandwidth is 90 nm at -1 dB, and the signal gain is up to 12 dB. The wide bandwidth, flattened gain, and low noise of QDFA are advantageous over the traditional erbium doped fiber amplifiers. The QDFA presented in this paper provides a new solving way of the growing demand for the bandwidth of fiber amplifiers in the current dense wavelength division multiplexing(DWDM) system.

The measurement precision of radial velocity needs improving for the detection of terrestrial planets. However, the spectral line shift during the change of coupling resulting from the insufficient scrambling of circular fibers limits the further improvement of the precision. The polygonal core fiber can provide better scrambling properties. Based the built experimental apparatus for measuring the near and far fields of the fiber output, the scrambling performance of the circular, rectangular, square and octagonal fibers under different coupling conditions is characterized via evaluating the near and far field patterns. The intensity variation in the far field and the barycenter shifting in the near field are investigated, and the scrambling gain is obtained for each fiber. The experimental results show that the polygonal core fiber has better scrambling performance. Especially, the octagon fiber has the minimum intensity variation in the far field, the minimum barycenter displacement in the near field, and the best scrambling performance. It is shown that replacing a circular fiber with an octagon one will effectively decrease the spectra line shift from coupling error and improve the measurement precision of the radial velocity.

The effect of the chaotic laser power and the single mode fiber length on the linewidth of Brillouin backscattering Stokes light is analyzed. It is found that along with the growing of input chaotic laser power, the Stokes linewidth decreases gradually and then nearly remains unchanged when the input chaotic laser power reaches a certain value. The effect of the input power and fiber length on the backscattering light power is analyzed. It is found that under the same input chaotic laser power, the backscattering light power grows rapidly with the increase of the fiber length and then tends to be saturated when the fiber length is beyond approximately 15.41 km. The influence factors of the chaotic stimulated Brillouin backscattering light threshold are also studied. The experimental results show that the chaotic stimulated Brillouin backscattering light has a high threshold, which is approximately 19 dB higher than that of the conventional continuous-wave laser.

Bessel beam generated by lens-annular slit system is analyzed in discrete Fourier transform method. In the theoretical part, Kirchhoff diffraction integral formula and Fresnel diffraction integral formula are expressed in the form of Fourier transform, the complex amplitude distribution of each surface is described in Fourier method. After collating and sampling, a discrete Fourier formula is derived to describe the intensity distribution of observation surface which has function relation with diffraction surface. After setting related parameters and simulating with MATLAB, the results show the intensity distribution of Bessel beam in different propagation distances. At last, an experimental system is designed to verify the theoretical analysis and numerical simulation. Simulation results and pictures from experiment show a higher match degree, which proves that this discrete Fourier transform method is feasible in lens-annular slit system.

Large number of sampling and low signal-to-noise ratio (SNR) limit the practical application of ghost imaging process. The traditional ghost imaging using Gaussian random field imaging ，however, speckle noise is introduced by the non-orthogonality of Gaussian matrix , which decreases the SNR and efficiency. Therefore, local Hadamard matrix is used to measure the target, and the clear image of low-resolution image is utilized as a prior knowledge for high resolution image to achieve denoising, which can effectively eliminate the correlated noise. Simulation and experimental results show that, comparing with the traditional reconstruction methods, the proposed method can increase the SNR and imaging efficiency.

In the course of studying the image quality assessment of photoelectric measurement equipment, the influence approach of the high frequency component of image spectrum to image quality subjective evaluation of the human eye and solving measures are analyzed. The clear image is processed by Butterworth low-pass filter in frequency domain. Then the influence of the high frequency part of image spectrum to the human eye subjective feeling is analyzed by filtering the image. The modulation transfer function (MTF) of defocus imaging system is measured by using the canted slit method, and the corresponding defocus blurred images are evaluated subjectively. Spectrum analysis and image evaluation of images acquired by the cameras with different pixel sizes are carried out. Experimental results indicate that the images have more high frequency component, richer details and higher clarity. Focusing exactly and capturing by camera with small pixel size are important measures to improve the image quality of the photoelectric measurement equipment. Other influencing factors and improving methods are proposed and the improving measures are put forward.

Image registration is the premise for extracting accurate inner element distribution information of samples from dual-energy computed tomographic (CT) images. In order to solve the misalignment of dualenergy CT images in synchrotron radiation, an image registration method is proposed which uses mutual information as the standard to measure similarity of two dual-energy CT images. The partial volume (PV) interpolation algorithm is used to calculate mutual information and the genetic algorithm is used to search the initial good parameters. Then the original searching point for Powell algorithm is initialized with these parameters and the PV interpolation based on Hanning window (HPV) is combined to search the optimal registration parameters. The numerical simulation testifies the advantages of HPV algorithm and the accuracy of registration achieves subpixel. Experimental results show that the proposed method is effective for dualenergy CT image registration and acquirement of three-dimensional distribution of elements.

To solve the problem of low resolution of catadioptric omnidirectional imaging, a system of catadioptric omnidirectional compressive imaging system based on coded aperture is designed. After the analysis of the relationship between point spread function and measurement matrix, the design of coded aperture pattern based on random point spread function is proposed. The coded aperture is programmably controlled with the relay lens and spatial light modulator. The high resolution of omnidirectional image is sparsely reconstructed based on omnidirectional total variation combined the characters of catadioptric imaging. The optic imaging platform is built to verify the effectiveness and practicability of catadioptric omnidirectional compressive imaging through the simulation experiment and actual equipment. The problem of low resolution of catadioptric omnidirectional imaging is solved effectively. The proposed method has important meaning in promoting the applications of catadioptric omnidirectional imaging in related fields.

Surface emissivity of casting billets is an important parameter that affects the temperature measurement on the billets. In this paper, an experimental apparatus is established by the method of radiation energy comparison. The apparatus consists of a heating system, an angle rotating system, a temperature detection system, a vacuum control system and a background radiation shielding system. The emissivity of GCr15 steel at different angles, temperatures and oxidations is measured by this apparatus. Research shows that emissivity increases first and then decreases with the increase of the angle. Emissivity increases with the increasing oxidation, and the angel of the maximum emissivity is gradually reduced. In addition, emissivity increases with the increasing temperature but the influence of the temperature is limited once the temperature is over 1000 ℃. At last, the measurement uncertainty is analyzed and the maximum emissivity uncertainty of the deep oxidized GCr 15 steel is 0.0205 when the temperature is above 1000 ℃.

A terahertz frequency comb system is built by using a commercial femtosecond fiber laser and a fiber coupled terahertz photoconductive antenna. The repetition rate frequency of the comb is highly stabilized based on phase-locked technique and the tracking stability is 3.3×10-13 at sampling time of 100 s. The beat signal between terahertz test source and terahertz frequency comb is achieved with the signal to noise ratio being superior to 50 dB. The absolute frequency of terahertz test source is measured based on the frequency comb method with measuring uncertainty of 5×10-11.

The measurement depth of traditional shadow Moiré profilometry is very limited. According to the space distribution of Talbot self-imaging under the illumination of point light source, a new technique using Talbot effect under point light illumination for shadow Moiré profilometry is proposed. An exact mathematical formula describing the relationship between the intensity of shadow Moiré fringes under the point light illumination and the distance between measured object and gratings is derived. The numerical simulation and experiment show that the measurement depth of the shadow Moiré profilometry is increased notably by using the Talbot effect compared with the traditional shadow Moiré profilometry. Respectively, the proposed method and the traditional shadow Moiré method are used to measure the surface of the same object, the difference of the results obtained by the two methods is within ±5 μm. Thus the validity of the proposed technique is proved. The proposed method can reflect the threedimensional morphology of object surface accurately.

In order to eliminate the phase jump error of existing phase unwrapping method based on multifrequency heterodyne principle, the improvement of the phase unwrapping method based on multi- frequency heterodyne principle is proposed. The partial fringes unwrapped phase is obtained by using the method of dual frequency heterodyne principle. Then the unwrapped phase is converted to other fringe relying on the relationship between the phase and the fringe pitch. After that, the wrapped phase is corrected by the converted phase, and the unwrapped phase is obtained. A series of constraints in the processing of unwrapping are derived in order to get the accurate and no jumped phase. The simulation and experimental results show that the method is simple and effective, and there is no phase jump in the unwrapped phase. The improvement of time efficiency is over 30% compared with the existing method.

Aiming at the existing problems in line detection by standard Hough transform, an improved algorithm based on Hough one- dimensional transform is proposed. Firstly, the edges of images are detected and the disconnected edges are grouped. Then, the Hough one-dimensional transform is adopted to detect in line and group the edges so as to reduce computation complexity and space complexity. At last, the problems on over-connecting, endpoints determining and detection precision are solved by added accurately processing. The experimental results prove that the improved algorithm can effectively solve problems in the standard Hough transform, reduce missing detection and false detection, improve the detection accuracy and achieve detection in line for different content images.

Wavelet ridge extraction is a key step in the wavelet transform profilometry. The speed and noise suppression capability of the wavelet ridge extraction method directly influences the wavelet transform profilometry. One new wavelet ridge extraction method used in the two-dimensional wavelet transform profilometry is proposed based on a new evaluation function. This function is established by using the module information from the wavelet transform coefficients, fringe instantaneous frequency and local fringe direction to improve the noise suppression capability. Moreover, one fast dynamic optimization algorithm is adopted to improve the wavelet ridge extraction speed. Computer simulation and experimental demonstration verify the proposed method’s effectiveness.

In order to improve the light fusion speed in multi-projector display system, a multi-projector display fusion method based on modulation information of structured light fringe is proposed. In this method, camera is used to capture the structured light fringes on the display wall which are projected by projectors, and then the modulation information for every projection channel is calculated. Meanwhile, the sub-pixel mapping relationship between projector and camera can be built by the phase of structured light fringes. With the above relationship, the modulation value for every pixel on projector can be obtained. Using both the modulation value and the edge blending results, we can generate the final sub-pixel fusion mask for each projection channel. The proposed method has the advantages of convenient operation, fast measurement speed and free of ambient light interference, and it has a wide application field. The theoretical analysis and real scene experiments both validate the effectiveness and feasibility of the proposed scheme.

A method to obtain three-dimensional (3D) trajectory of moving target through single X-ray source imaging is presented. 3D trajectories of moving objects are obtained through target detection, stereo matching and 3D reconstruction based on the digital radiography image array collected. Compared with common approach of multi-view imaging, this method has advantages of simple device and equivalent accuracy. Simulation shows the validity and stability of the proposed method. The error analysis of 3D trajectory measuring result is also given. It turns out that the measuring error is related to the detector resolution and target size.

The Slot structure is superior in improving the sensitivity of integrated optical waveguide sensor and decreasing the detection limit. The microring optical biosensor based on PSQ-Ls polymer waveguide with the slot structure is studied. The influence of height, width and slot width of the Slot waveguide on the sensitivity is analyzed at the 850 nm band, and the optimized cross section parameters of microring sensor are gained for the single mode transmission. The bend loss and free spectral range of the Slot waveguide microring are analyzed, and the relationship between structural parameters and performance of microring sensor, such as extinction ratio, quality factor, is obtained. The optimized bend radius and coupling efficiency of the microring are achieved. Compared to the ridge waveguide based microring sensor, the Slot waveguide based microring sensor has twice of the sensitivity and a half of the detection limit.

Frequency doubling in an external cavity is a prevalent method to generate an ultra- violet laser at 397.5 nm . A semi-monolithic resonant frequency doubling cavity based on the PPKTP crystal is built and is used to realize the resonant frequency doubling of the 795 nm single frequency continuous-wave laser amplified via a semiconductor tapered amplifier. Under the condition of 203 mW input power of a 795 nm laser, the 397.5 nm single frequency continuous-wave ultra-violet laser with a power of 60.4 mW is obtained, and the frequency doubling conversion efficiency is 30%; and the maximum doubling efficiency is 34.6% with a fundamental power of about 87.5 mW. The beam quality factor M2 of the frequency doubling ultra-violet laser is superior to 1.21, indicating the better beam quality. The typical root-mean-square fluctuation of the output power within 30 min is less than 1.9%. This frequency doubler is compact, has good mechanical stability, and can be used to achieve steady output of ultraviolet laser. The ultra- violet laser can be used to generate the squeezed or entangled states of the rubidium transitionline, and plays an important role in the quantum optics, precise measurement and other fields.

A high power level ytterbium doped all-fiber laser at 1018 nm is proposed using the resonant cavity with the 125 μm - diameter double cladding fiber. The single mode high output power of 254 W is achieved with the optical-optical conversion efficiency of 81% via optimizing the length of the gain fiber and controlling wavelength of the pumping LD. The amplified spontaneous emission (ASE) is suppressed efficiently and the residual pumped light is dissipated completely, while the signal to noise ratio (SNR, RSN) is more than 35 dB. As the wavelength of the pumping LD changes with the output power, thus affecting the conversion efficiency of the fiber lasers, the wavelength of pumping LD needs to be controlled accurately for obtaining more efficient conversion efficiency. The output power instability is measured in the experiments with the value of less than 0.5% in four hours. The high efficient fiber lasers at 1018 nm with the stable output power can be applied to the tandem pumping high power fiber lasers as the excellent pump source.

In order to reduce the mismatching rate of binocular stereo matching algorithm in the disparity discontinuity region and under noise disturbance, a stereo matching algorithm based on improved Census transform and dynamic programming is proposed. An improved Census transform with a noise margin is applied to compute the cost based on a cross shape support region. The reliability of single pixel matching cost is enhanced. The guided image filter is used to aggregate the cost volume fast and efficiently. In the disparity selecting step, an improved dynamic programming algorithm is designed to eliminate the scan-line effect and improve the matching speed and accuracy. The final disparity maps are gained after post-processing. The experimental results demonstrate that the proposed algorithm evaluated on the Middlebury benchmark achieves an average error rate of 5.31% , and the accurate disparity can be obtained in both low texture and disparity discontinuity regions with low computing complexity and strong robustness.

Laser tweezers Raman spectroscopy and singular value decomposition are used to explore the synthesis dynamics of biodegradable plastic poly- β- hydroxybutyrate(PHB) in C. necator H16 strain cultured with different nitrogen sources at the single- cell level. The results show that ammonium sulfate performing best on fermentation speed, efficiency and productivity in all tested nitrogen sources, is demonstrated as an ideal nitrogen source for PHB fermentation. The singular value decomposition results reveal that the main fermentation features are the Raman peaks belonging to RNA, DNA, protein and PHB, and the difference in bacterial cells and product yield becomes more notable as the fermentation process works. The dynamics of Raman intensity peaks of nucleic acid, protein and PHB at 782, 1574, 1660, 1732 cm-1 are active to the fast PHB synthesis, and no matter what nitrogen sources are used, the intensity of peaks at 1660 cm-1 and 782 cm-1 is positively correlated, but the intensity of 1660 cm-1 and 1732 cm- 1 peaks is negatively correlated. The results suggest that nitrogen sources influence the RNA and the protein metabolism and then affect the synthesis of PHB indirectly. Raman spectroscopy combining with data analysis can provide the metabolism information of the microbial fermentation and offer a powerful experiment basis for optimizing fermentation process.

To overcome the low luminous flux in separated portable Raman spectrometer, an optical system embodies the probe of Raman spectrometer and monochromator is designed. An aspherical lens with high numerical aperture is used to obtain the effective excitation of sample and the efficient collection of signal in the optical system of the probe. A cemented lens group is used to minish the size of converging path and eliminate chromatic aberration. The monochromator is based on asymmetrical crossed Czerny-Turner structure. For the purpose of achieving desired spectrum resolution and spectrum range, the relationship between the spectrum resolution and spectrum range of the monochromator and the structural parameters of the asymmetrical crossed Czerny-Turner system is established. According to the tested spectrogram of mercury lamp, the spectrum resolution of the monochromator is better than 6 cm-1 (0.37 nm), and spectrum range is 790~950 nm (200~2000 cm-1). A sample of CCL4 is introduced to examine the performance of the optical system, the experimented results demonstrate that the peak intensity of spectrogram of CCL4 measured by this incorporative optical system is almost three times as high as that using a commercial probe couples to the monochromator, which verifies the rationality of the design of the optical system.

A new multi- plane mirrors linear combination solar concentrator is proposed for concentrating photovoltaic system. Based on the new concentrator, the geometric concentrator parameters that affect the optical performance are studied using analytic geometry method. The concepts of the minimum design spacing and cosine efficiency are proposed. The interactional condition between focal length and different sizes of sub-mirror, as well as between focal length and different numbers of sub-mirror investigated, and the optimization of the minimum design spacing and cosine efficiency are given. A model developed by TracePro software is used to obtain the flux distribution of the focal plane at different simulation parameters. An experimental prototype concentrator with 450× suns is fabricated and preliminary tested, the operation shows that the effect is obvious and the expected results is arrived. The simulation results can present a useful reference for design and optimization of the solar concentrator.

To improve on the lamp source set of the system for the heat flux sensor calibration, the simulation study is implemented in the paper. Based on the luminance distribution, the composition of xenon short arc lamp is designed and its power distribution is calculated. The angle distribution is simulated by angular apodization files. Then the model of the lamp source set is built and the radiation on the exit of the integrator is calculated. Compared with the test result, the average radiation on the exit of the integrator in test deduces about 49% and the loss of radiation increases with the increasement of the caliber of the integrator. By theoretical analyzing and experimental verifying, the loss of radiation is because the collection efficiency of the ellipsoidal specular reflector falls by 40% in test. And the reason why the loss increases with caliber of integrator is that the xenon short arc lamp model in this paper reflects well the feature that the density of the power of xenon short arc lamp descends from the cathode to the anode.

High precision lithographic projected objective suffers from thermal aberrations due to energy absorption. Under off-axis illumination, the dipole mode and time-changing thermal astigmatism is outstanding, which cannot be corrected or compensated by traditional passive optical systems. Active optics using mechanical actuators to compensate the primary thermal astigmatism of deformed lens in a refractive lithographic objective system is proposed. The actuators are located on a refractive plate, and the finite element analysis is applied to study the deformation and aberration. The theoretical analysis by geometrical optics is made to prove the feasibility for compensation, and the influences of actuator locations, active force loads, actuator size and lens supporting conditions are considered. The results show that the active plate performs well to compensate primary astigmatism and primary 4- foil aberration under the optimized supporting conditions, which provides an idea for thermal aberration control in lithographic systems.

An optical solution to boresight error using tilted optical elements is presented. It can correct dynamic boresight error for conformal windows which vary with look angle. A model of boresight error for conformal windows is established, and dynamic boresight error introduced by a conformal window is calculated. The principle of boresight error correction of conformal optical windows with titled optical elements is stated, which is proved by adding a tilted spherical lens behind a conformal window. A design example is presented, and by tilting the corrector which is used to correct the conformal window aberration, the boresight error for the conformal window is corrected while the aberration correction for the window is not hindered.

A compensation method based on radial preload is presented in order to decrease the surface deformation of a precision mirror caused by gravity. By taking the mirror used in a high numerical aperture (NA) lithographic objective lens as a research object, a force analysis model is established. The influence law of surface deformation affected by the radial preload is analyzed qualitatively under the action of gravity. Surface deformation of the mirror is obtained under the condition of different preloads by the finite element method. The relationship between radial preload and surface deformation is analyzed through data fitting as well as the relationship between the radial preload and the Zernike coefficient. The analysis results show that the radial preload mainly influences the spherical aberration and the trefoil aberration of the mirror surface, and the compensated surface deformations of the mirror first increase and then decrease with the increase of radial preload. The root mean square value of the mirror surface deformation decreases from 2.009 nm to 0.462 nm, which is the optimal compensation result. Moreover, the mirror surface deformation is measured experimentally under the action of gravity and the radial preload, and the experimental results indicate that the mirror surface deformation caused by gravity decreases by a mean of 0.988 nm when the radial preload is about 25 N. The validity of the analysis process and the compensation method are confirmed.

Based on the vectorial Debye theory, the tightly focusing properties of Bessel-Gaussian radially polarized beam through a high numerical aperture (NA) with a diffractive optical elements (DOE) and a dielectric interface are investigated. A three-dimensional optical chain along optical axis can be obtained by designing the DOE to spatially modulate the incoming Bessel-Gaussian radially polarized beam. Research results show that the focusing shape in the focal area can be controlled by changing the interception ratio of the aperture radius to the incident beam waist radius, numerical aperture and the refractive index of the medium in the focal region. When the interception ratio is beyond to a constant value, a focal dark channel can be obtained. The length of the focal dark channel increases with the increasing of refractive index of the medium in the focal region in case of not changing the width of the focal dark channel.

Subtracting a photon from a squeezed vacuum state is an efficient method for preparing an optical Schr?dinger cat state. It is shown that the fidelity of the Schr?dinger cat state based on photon subtraction depends on the purity of squeezed vacuum state. A two-mode squeezed state of light with a purity of 0.993 is experimentally produced by a non-degenerate optical parametric amplifier, where the pump beam oscillates partially in the cavity. Especially, two-mode squeezed states with high purity can be obtained by the non-degenerate optical parametric amplifier in a wide range of pump power due to the presence of thermal effect.

The dynamics of two qubits with inter-qubit interaction is investigated by an external quantum oscillator in its coherent state. The coupling between the two qubits and the oscillator is in the ultra-strong region and their detuning is large. With adiabatic approximation method, the reservation of one initially entangled state of two qubits by the quantum oscillator is accomplished through the careful choice of the four parameters, the average number of the coherent state of oscillator, the ultra-strong coupling strength, the ratio of two frequencies of qubit and oscillator, and the inter-interaction coupling of two qubits. If the inter-qubit interaction strength is negative, their entanglement can be kept further for a much longer time. These results are different from those in the previous studies and reasonable explanations are given. Maintaining the entanglement of two qubits with very high fidelity will be helpful for the quantum information process.

A classification method based on deep learning is proposed for the classification of high spatial resolution remote sensing images. The texture features of the images are calculated through nonsubsampled contourlet transform, the deep learning common model- deep belief networks (DBN) are used to classify the high spatial resolution remote sensing images based on spectral and texture features. The proposed method is compared with the DBN classification method based on single spectral information, the support vector machine (SVM) method and the traditional neural network (NN) classification method. Experimental results show that comparing with the single spectral information, the use of spectral and texture information can effectively improve the classification accuracy of high spatial resolution remote sensing images, and comparing with methods of SVM and NN, the DBN method can accurately explore the distribution law of the high spatial resolution remote sensing images and improve the accuracy of classification.

In order to measure the on-orbit modulation transfer function (MTF) of medium resolution spectral imagers, an approach based on non-ideal step targets is proposed. By using targets with only one uniform side, the proposed method based on the step response is carried on the assumption that the line spread function (LSF) is symmetric. Compared with the traditional step edge method, this method achieves uniform results from the same target captured by a medium resolution spectral imager. The full width at half maximum (FWHM) derived by the proposed method differs from the traditional methods by only 0.03 pixel and the difference of the MTF value at the Nyquist frequency is only 0.02, which verifies the effectiveness of the proposed method. Since the using targets are very common in remote sensing images captured by medium resolution spectral imagers, the proposed method can be applied to more cases than the traditional approaches.

The expression of Hermite-Gaussian beam in terms of spherical vector wave functions is given with the complex source point method. Based on the generalized Lorenz-Mie theory, the scattering coefficients are obtained by applying the continuous boundary conditions of tangential components of electromagnetic field at the interface between chiral medium and free space. The scattering of Hermite-Gaussian beam by a chiral sphere is investigated. The angle distribution of far region scattering field by a chiral sphere located in the Hermite-Gaussian beam is numerically simulated. The influences of beam mode, chiral parameter and sphere radius on the scattering properties sections are analyzed in detail.

Vanadium oxide films (V2O5) are deposited on the substrates of infrared quartz, sapphire and singlecrystal diamond respectively by radio freguency (RF) reactive magnetron sputtering. The structure, thickness, surface morphology, electrical and optical performances of the films are studied. The results indicate that all the deposited films are polycrystalline V2O5 films with preferred orientation along (001) plane. The crystallization behavior and the surface topography are the best when the substrate is single-crystal diamond, the phase-transition temperature and the temperature range are the lowest simultaneously. In terms of the optical performance, the optical-shutting time of the films on sapphire and diamond is less than 2.5 ms and the recovery time is about 30 ms . The prominent optical-mutation property is presented on the single-crystal diamond, with the transmittance ratio before and after the phase transition of 10.3.

A design of four- layer broadband antireflection coating (ARC) for high efficiency GaAs solar cell [inverted metamorphic 3 junction (IMM-3J),4 junction (IMM-4J) lattice-mismatch solar cell] is proposed. The broadband antireflection coating is prepared on the solar cell by electron- beam thermal evaporation method. According to the differences of reflectance and external quantum efficiency (EQE) before and after evaporation, the short circuit current gain of each junction is calculated. Compared with the double-layer antireflection coating, the four-layer antireflection coating with a wide anti-reflective area is more appropriate for IMM-3J and IMM-4J solar cell. Process stability of the antireflection coating preparation is analyzed by Essential Macleod software. The related design optimization is also given out.