A confocal laser scanning endomicroscopy is developed, which employs a telecentric endomicroscopic optical system to achieve long-distance relay transmission of images, telecentric f-theta optical scanning and endomicroscopic imaging functions. Two-dimensional confocal scanning is realized through two scanning mirrors, and low-noise control signals are generated by the embedded system. Miniaturized design is adopted in the confocal laser scanning endomicroscopy in order to easily use for portable applications. Firstly, the outer diameter of the optical system is 8 mm, and working length of which is 250.3 mm, so in-vivo endomicroscopic imaging can be realized through standard laparoscopic surgery port. Secondly, two small mirrors with aperture of 3mm are employed to achieve in vitro confocal scanning, with the maximum frequency of 100 Hz. Finally, laser system and fluorescence detection are connected with the distal end of the confocal laser scanning endomicroscopy through an optical fiber and a cable, and the size and the weight of the distal end are minimized by this method. The experimental results show that the system′s filed of view reaches φ600 μm, and its optical resolution is 2.2 μm. This system can be used under handheld operation or other ways to realize minimally invasive in-vivo fluorescence endoscopic surgery since its miniature size and low weight.

In order to solve the contradiction between the disc size and resolution, and to attain the high response speed, a new photoelectric absolute angular encoder is proposed based on a single-track coded disc and a linear detector. A new single-track absolute coded disc is used in the encoder. The encoded units are directly carved on one track. The image of coded disc is magnified by the imaging lens, and received by the linear detector. The FPGA control circuit sends the data of image to the computer for data processing. An integrated 7 bits binary code is obtained by image processing and the subdivision code unit is gained by pixel subdivision technique respectively, which are both contributed to the angular information. An absolute angular encoder, which contains a diameter of 40 millimeter code disc, is designed based on the technique. Its resolution can reach 15 bit.

In order to solve the problems of large error and uncontrollable clearance caused by optical de-rotation gear transmission in panoramic electro-optical observation and sighting equipment, a position control system is proposed in which the optical de-rotation mechanical transmission is removed and substituted with a position control system without mechanical transmission.Using modern control theory and on the basis of analyzing the subject model in the optical de-rotation position control system,the complex control algorithm is designed, dominated by Proportional-Integral-Differential(PID) controller and aided by nonlinear control. The position control system is established,adjusted and tested with the single chip processor C8051 served as the core of controlling,the coder as the position transducer,electro-optical stage azimuth as the controlling singnal.The test results indicate that the optical de-rotation performance is better than that of the mechanical transmission, and the high precision position control is realized.

The yield of curve-shape arrayed waveguide grating (AWG) chips is twice of the square-shape ones. However, the center wavelength of curve-shape AWG is easily influenced by the packaging stress, which is introduced by the linear expansion coefficient difference between the packaging box and the ribbon fibers coupled to the AWG chip. The experimental results showed the AWG center wavelength varied with the thermal stress linearly. The variation of the center wavelength for a thermally stabled AWG module was as large as 46 pm, when the ambient temperature varied from －20 ℃ to 65 ℃, even the ribbon fibers were bonded to the packaging box with soft silicon rubber. In this paper, a new low stress thermal stabled AWG packaging was introduced. In this package, a borosilicate glass plate was bonded to the ceramic heater, and the ribbon fibbers were bonded to it with silicon rubber. For the linear expansion coefficient difference between the borosilicate glass plate and ribbon fiber is very samll, the thermal stress of ribbon fibber or AWG chip is very small. Experimental results showed the typical center wavelength variation for thermally stabled AWG modules manufactured with this technique was smaller than 5 pm, and the typical thermal stress was smaller than 0.029 MPa, when the ambient temperature varied from －20 ℃ to 65 ℃.

Multimode glass planar waveguides were fabricated by K-Na ion exchange technique. The effective indices were measured by prism coupling technique. The refractive index profiles of the planar waveguides were reconstructed by Inverse Wentzel-Kramers-Brillouin method and the best fitted refractive index profile is Gaussian distribution. The fabrication condition of single mode waveguide was got from the dispersion curves and a single mode waveguide were fabricated. The surface refractive index obtained by WKB dispersion equation. The image of propagation in the waveguide was taken by digital camera, from which the light intensity attenuation curves were fitted and the propagation losses were calculated and the propagation loss of the single-mode waveguide is about 0.4 dB/cm.

A three-dimensional optical switch is introduced which is composed of two N×N three-dimensional multi-mode interference couplers and an array of phase shift waveguide. Firstly, the general imaging principle of three dimensional multimode interference is analyzed by the guided-mode propagation method，and the imaging position and phase matrix are deduced. And the transmission equation of optical switch is set up by the transmission matrix method.The phase condition when the optical switch works is calculated by the transmission equation,and verification and the simulation is completed by three-dimensional finite-difference beam propration method. The insertion loss and crosstalk when the optical switch works are given, and the relation between the output power and the phase shift of phase array waveguide are obtained.

Two-dimensional gap plasmonic waveguide employing surface plasmon polaritons is proposed. Transmission characters of 90 degrees sharp bend waveguide, T-shaped splitter and switch based on gap plasmonic waveguide are numerical simulated with finite-difference time-domain method. It is shown that the energy loss of 90 degrees sharp bend waveguide is ohmic losses which are different to radiation losses of dielectric optical waveguide. When the gap width is larger than 40 nm, the transmission of bend waveguide is larger than straight waveguide with the same length. In the T-shaped splitter, when the gap width ratio of two output waveguides is larger than 0.6, the output energy is mainly assigned to the output waveguide which effective index is smaller. The result is different from dielectric optical splitter. The T-shaped switch is off with two input light out of phase and is on with two input light in phase. The plasmonic modes in the above devices beyond the diffraction limit and can be employed in super-large-scale optical integrated circuit.

The experimental research of optical channel waveguides in Z-LiNbO3 crystals using femtosecond pulses is performed on Ti sapphire laser system, the wavelength and the repetition rate of it are 800 nm and 76 MHz. By using the focusing lens with 40 magnification, o.65 numerical aperture (NA), the effect of fabrication parameters, for example laser scanning speed, repeat time on writing quality and the characters of relative microstructures are analyzed and studied in theory. Also the distribution is measured by the beam profiler. The result shows that the propagation of optical waveguide could achieve ideal effect when the pulse energy is under 250 mW, the scanning speed is 100 μm/s and the focus depth is from 350 μm to 400 μm. At last the insertion loss of 1×2 optical splitter is about 1 dB/cm.

In the rotating-wave approximation, the produce and control for periodic quantum echo of atom bit interacting with a two-mode coherent light field is investigated via intensity-dependent coupling multiphoton channel by means of the full quantum theory. The effects on the evolution of the fidelity of atom bit for both the distribution form and distribution range of the average photon number of the two-mode coherent light field, and the photon number k of absorbed (emitted) as the atoms transition are discussed by numerical calculations, respectively. The system parameter of preparation and control for periodic quantum echo of atomic bits are obtained. According to the entanglement theory, the relation of the fidelity of atom bit with atomic reduced entropy is analyzed. The results show that atom bitkeeps good coherence and fidelity, and will produce periodic quantum echo for the two photon process as k=1, by regulation average photon number of the light fields being symmetrical or asymmetrical distribution and its value in a certain range. For k≥2 multiphoton process, the atom bit is always in a maximally entangled with the two-mode coherent light field so that the atom bit is in a state of partial distortion without periodic quantum echo. It is shown that the physical nature of the periodic quantum echo produce is periodic disentangled between the atom bit and the light field.

To improve the performances of the present Ping-Pong protocol systems over noisy channels, an efficient quantum secure direct communication scheme is proposed. Entanglement-assisted quantum low-density parity-check code is used to protect the encoded quantum information sequences against the channel noises, which improves the reliability of the quantum secure direct communication system. Furthermore, quantum automatic repeat-request protocol with error detection code is used to detect the strong channel noise effect, which improves the communication efficiency. Two steps of eavesdropping detection are taken to ensure the system security, and the iterative decoding performance of entanglement-assisted quantum low-density parity-check code is simulated over quantum depolarizing channel.

Fan filter palys an important role in direction detection of images. In this paper, a simple but efficient method is proposed for the design of two-dimensional (2D) fan filters with arbitrary orientations. The proposed method is based on the polar Fourier transform (PFT) and a wedge-shaped filter. Firstly, the wedge-shaped filter is transformed to the polar Fourier domain. Then, by taking the rotation advantage of PFT, this wedge-shaped filter is rotated via shifting its PFT along the horizontal axis. By doing so, a series of arbitrarily-orientated fan filters are obtained and these obtained fan filters can detect the image information in arbitrary directions. Furthermore, since the design process involves no 2D optimization, the advantage of easy design is also achieved. To varify their directional sensetivity, the designed fan filters are applied to detect the texture direction of images. Experiment results show that the proposed fan filters has great potential in texture detection.

The general segmentation algorithms are mainly dependent on visual features extracted from images, such as intensity, color and texture features. However, the feature extraction from hazy image becomes difficult and unstable. In this paper, a novel feature vector suitable to hazy image segmentation is introduced, and the corresponding feature extraction algorithm is proposed. The feature vector is composed of three components, which are degree of polarization of objects, depth map and color information, respectively. The degree of objects is separated with the degree of airlight using an decorrelation-based method. The depth map is inferred by analyzing the polarization images based on the hazy degradation model. And the color information is obtained from non-polarization image estimated by polarization images. Finally, these features are formed into a vector and fed to a well-known graph-based segmentation algorithm. After comparing the proposed results with those using only color cue on two aspects, it can be concluded that the proposed feature vector is more effective and robust than the usual features for hazy scene segmentation.

A maximum likelihood estimation image denoising method is proposed using the non-local principle component analysis. Pixels with high similarity in both the gray level and the texture information are selected, and used to implement the maximum likelihood estimation. This kind of optimal restored method can overcome the drawback of the local image denoising method such as blurring edge, and improve the accuracy for restoring detail information in image using maximum likelihood estimation. Experiments using the proposed method, principal neighborhood dictionary non-local mean method and local maximum likelihood estimation method are implemented in images with different noise standard and different geometric complexity, and the performance of aboving three denoising methods are compared quantitatively and qualitatively. The results demonstrate that the proposed method can remove the noise effectively and preserve detail imformation of images compared with the currently used methods.

Errors of two Inverse Perspective Mapping (IPM) formulae are analyzed in mathematics,and a new IPM formula is deduced.An experiment is designed to validate the IPM result from different formulae.Experimental results show that,whatever the parameter of camera is,the IPM result obtained by the third formula has no distortion.In the special condition of camera installation,the average error of the deduced formula is 0.85%.Consequently,the IPM transform result of the improved formula is more precise and can satisfy the system requirement in smart navigation and road surface reconnaissance field.

For the fast calculation of the short time Fourier transform,the AFDISAR signal can be preprocessed by the proposed method to get a range of the speed of the slapper,and a range of the wavelet transform scale factor can be estimated.Then,the signal is analyzed by continuous wavelet transform method.The processing results of simulated interference signals suggest that the relative error is less than 2%.The displacement interferometer in high-speed detonation test is analyzed separately by method of short time Fourier transform,of wavelet transform and the method of short time Fourier transform combining with wavelet transform.The third method can recover the velocity history accurately,and calculate error is in the tolerated rang of the system.Taking calculating speed and calculating accuracy into consideration,the third method is chosen to the AFDISAR signal analysis.

In the phase shift modulation type optical components constructed by traditional dielectric waveguides with the low refractive index constant based on the total inner reflection mechanism, the length of the modulation region is usually millimeter or centimeter order of magnitude, while their horizontal size is micrometer of order of magnitude, therefore, the most typical characteristics of optical waveguide devices are long and narrow. Reducing the size of optical waveguide devices is a hard problem in the development of highly dense integrated optical circuits. The emergence of photonic crystals provides a new approach to develop highly dense integrated optical circuits. The plane wave expansion method is used to calculate the dispersion curves of the line-defect type photonic crystal waveguides. It is observed that there is a large change of wave prorogation constant near the guided-wave band edge, corresponding to a little change of refractive index of the material. If the work frequency is selected near the guided-wave band edge, the phase shift modulation length is expected to largely reduce. The finite-difference time-domain method is used to demonstrate the results above. Calculated results indicate that there is strong enhancement effect of refractive index phase shift modulation near guided-wave band edge: for the refractive index change of 0.46%, the phase shift modulation length in these waveguides is only 11.7% of that in conventional uniform dielectric material. This enhancement effect is originated from the special flat dispersion properties near the guide-wave band edge, and it is expected to be applied to high dense photonic integrated circuits after further research.

Grating structures of gold nanowires with a width of sub-100 nm and a total area in the order of square centimeters are prepared using interference lithography and colloidal gold nanoparticles.In the fabrication, photoresist is firstly spin-coated onto the glass substrate that is coated with a layer of indium tin oxide as thick as 200 nm.Grating structures are then produced by interference lithography into the photoresist.Through controlling the exposure dose and the development time, small duty cycles are achieved with the photoresist gratings.Thereafter, colloidal gold nanoparticles are spin-coated onto the photoresist master grating.Making use of the dewetting properties of the colloidal solution on the photoresist surface, very limited amount of gold nanoparticles remain on the photoresist grating and are confined into the grating grooves.In combination with the small duty cycle of the photoresist grating structures, this mechanism enables realization of narrow gold nanowires.In the final stage of the fabrication, the sample is heated to 250℃ for about 5 minutes.Gold-nanowire gratings with a duty cycle of 1: 4 and a period of 400 nm are achieved, which have an effective area of 1 cm2.The optical response featured with strong coupling between the waveguide resonance mode and particle plasmon resonance confirms the successful fabrication of the waveguided metallic photonic crystals.This device may be utilized as the central part of the biosensors with small sensing volumes.

To meet the requirement of market, a short focal digital projection lens for 0.8 inch digital light processing(DLP) projectors is designed. The total length is 172.6 mm, and the full aperture is 70 mm. The lens uses a retrofocus structure, and it is composed of six glass lenses and one plastic lens (two even aspheric surfaces). Its FOV is 80°, F-number is 2.1, retrofocus proportion is 3.17∶1, and projection ratio is 0.76∶1. Its MTF in all field is less than 28° that is higher than 0.6 at the limiting spatial frequency of 37 lp/mm. The lateral color is less than 3.5 μm. The distortion of the full of view is less than 1.1%. Its image quality is excellent with advantages of simple structure, small size, easy process and low cost.

Electromagnetic properties of single-layer silver nano-particles of compound metamaterials are theoretically analyzed. MATLAB numerical calculation and CST simulation are used to verify the theory. The results show numerical and simulated results of 4.75 nm nano-silver arrays are basically the same, and there are Lorentz resonance. The simulated results have redshift compared to numerical results, having a weak magnetic resonance. Further simulated compounded by the use of 4.75 nm spherical to form large particles of 100 nm, the results show that there are negative real permittivity and negative real effective permeability and negative refractive index in the 300~380 THz range. That means 100 nm composite spherical particles have a strong magnetic resonance effect and consistent with the theoretical analysis.

Aiming at the problem of undersampling in the aspheric surface point diffraction interference measurement, a method for solving the location of the center of the best fitting spheric surface is proposed, to reduce the interference fringes density. The method is based on the minimum reflex convergence spot radius. Firstly the important role that obtaining the location of the center of the best fitting spheric surface plays in the high-precision aspheric surface measurement is analyzed. And taking minimum reflex convergence spot as the criterion to judge the location of the center of the best fitting spheric surface is pointed out. Then the computing model for the location of the center of the best fitting spheric surface is derived with ray tracing and iterative. Also the locations of the center of the best fitting spheric surface based on different asphericity definitions are compared. And the effects of aspheric surface parameters to the locations of the center of the best fitting spheric surface and the convergence spot radius are analyzed. The results indicate that, the best fitting spheric surface parameters based on minimum convergence spot radius is suitable for the aspheric surface point diffraction interference measurement model, also can be used to guide the assembly and debugging of the interferometer and the selection of wavefront reconstruction parameters.

An intelligent temperature compensating circuit is used to compensate the avalanche photodiode reverse bias voltage for offsetting the impact of environmental temperature on the avalanche photodiode, thus reducing the temperature drift of the system.Using this technology, the system measurement error caused by ambient temperature drift can be controlled within ±0.1℃ from 0℃ to 60℃.Compared with traditional constant temperature devices, using temperature compensating circuit can effectively reduce system power consumption and cost.This circuit has better temperature compensation linear, and is more flexible for compensation coefficient setting than the circuit using thermistor.

In order to accomplish the all-optical control in the aspect of tempaerature sensors, a theoretical design of temperature sensors is proposed based on the nonlocal spatial solitons. Taking lead glass as the propagation media of the soliton, the phase shift of the center-incident soliton is studied in the cylindrical lead glass. It is concluded that a temperature change of 0.32 ℃ can lead to π phase shift of the nonlocal solitons. The phase shift modulation by the temperature change will be a practicable and sensitive design of the temperature sensor based on the interferometer theory.

According to the fiber Bragg gratings optical sensing principle,a novel structure of fiber Bragg settlement sensors are developed,based on a cantilever beam and a metal diaphragm.The sensing characteristics are investigated.In the experiment,water level differences to are used simulate foundation settlement.The experimental results indicate that sensitivity of -2.11pm/mm and good linearity of higher than 0.999 can be obtained.By changing the thickness and the effective length of cantilever beam,the measurement range and sensitivity of the sensor can be optimized to meet various applications.From the results,it is known that the sensor will play an important role in bridges,railways and other foundation settlement monitoring.

A refractive-index sensor with micro-structured defect was proposed and demonstrated,and its refractive-index characteristics were studied experimentally.After inserting a segment of thin-core fiber into a standard single-mode fiber,high-order cladding modes were excited and interfered with the core mode to form an in-fiber modal interferometer because of the mismatch between thin-core fiber and standard single-mode fiber .By decreasing of the cladding-diameter of the thin-core fiber,the evanescent field of the cladding modes was enhanced,which improved the sensitivity of refractive-index.Experimental results show that the proposed device has advantages of low loss,low cost,high sensitivity and good linearity.

Because the diameter of the trapped sphere obtained by the optical tweezer usually locates in the intermediate range, theoretical calculation was difficult. OTT1 optical tweezers computational toolbox is a kind of T matrix method based on the GLMT theory. The development of the OTT1 optical tweezers computational toolbox makes detailed calculation and evaluation feasible. In this paper, the optical trapping coefficiency curve, its linearity and stiffness, and the interchangebility of Duke series polystyrene sphere were calculated and evaluated. The results show that when the numerical aperture becomes closer to the refractive index of water, a better trapping effect will be achieved; when an 800~1 200 nm polystyrene sphere is trapped, the stiffness of optical tweezers is larger; when a PS sphere with diameter smaller than 2 000 nm is used, the linearity of optical tweezers is better; when Duke series sphere with a 800~2 000 nm diameter is adopted, the interchangeability between the same series spheres is better, which will be convenient to calibrate the nanometer optical tweezers; for avoiding Mie resonance, the sphere diameter should be smaller than 2 500 nm.

The traditional laser range finder cannot meet the need of real-time and the measurement accuracy. Owing to this, a multi-frequency modulation laser range finder system was designed based on orthogonal calculation and CORDIC phase solver. An improved orthogonal algorithm was used to calculate the tangent of the ranging phase for noisy environments, and the angle was calculated by CORDIC angle solver algorithm. The proposed algorithm effectively improved the range accuracy and greatly reduced the computational. In experiments, the sampling frequency of AD converter is 500 MHz, the SNR is 14 dB, the word length is 16-bits, the measurement range is 150m, the phase difference resolution is better than 0.026 4° and the distance accuracy is better than 0.11 mm.

Aiming at the shortcoming that the M-th power based traditional phase recovery method cannot be used for high-order QAM signal constellations, a carrier phase recovery method for 16-QAM constellations is presented. The proposed method is able to applied in CO-OFDM system to implement blind phase recovery with low computational cost. In the verification experiment, the method is used in 112 Gb/s 1 040 km CO-OFDM transmission, and its performance is compared to those using the fourth-power estimator. Experimental results indicate that the proposed method significantly outperform the fourth-power estimator.

A probe of the common-path optical coherence tomography (CP-OCT) is presented based on the wedge-shaped and angled fiber, which can achieve the phase interference for the backscatter signal, so that the cross-section imaging information of the sample under test can be obtained.Firstly, an optical path of the CP-OCT and its theoretical model of phase interference at the probe end are set up, and the relationship between probe size and interference strength is analyzed.Secondly, the optical fiber probe with the wedge-shaped end is designed, and the experiment system for the CP-OCT at the wavelength of 1550nm is built.And then, depended on the scattering rate of the sample, various probes with different angles are prepared and the system performance with different probes is analyzed.The experimental results show when the end of the probe is cut to a certain angle shape, the output splitting ratio can be adjusted to achieve the adjustment of the reference signal so that signal noise ratio can be improved.The research works lay the foundation for the establishment of the effective CP-OCT system.