A colloidal gold lateral flow (LF) strip reader based on CMOS image sensor, characterized by ultrasensitive and high repeatable, is developed. As gold nano-particles absorb green light strongly, green LEDs are used as the light source in the reader. In order to get the best illumination uniformity, gray value distribution of blank strip image is used to adjust the location and intensity of LEDs. CMOS control parameters are specified to get the most suitable image quality for image analyzing by self-defining parameters. Blank strip is used to calibrate the background signal. The quantitative result is derived from the ratio of integral reflection optical density of T line signal to C line signal. LF strips of cardiac troponin I (cTnI) with standard mass concentration gradients from 0.25 ng/ml to 64 ng/mL are detected. The 4-PL quantitative model is used to fitting standard working curve, whose correlation of coefficient (R2) is more than 0.99. The coefficient of variation (CV) of 20 times repetitive experiments of 0.25 ng/mL is 2.790%, and 64 ng/mL is 0.134%. The functional sensitivity (FS) is better than 0.25 ng/mL. The reader is sensitive, stable, simple, and suitable for field detection.

Raman spectroscopy for studying of chemical vapor deposition (CVD) diamond is one of the most effective means. CVD diamond films with different grain sizes are analyzed by Raman spectrum. Ultraviolet Raman spectrum test shows that ultraviolet stimulated Raman spectrum effectively restrains the scattering of sp2 carbon and the non-diamond fluorescent background. The presence of 1140 cm-1 peak in the visible Raman spectrum attributed to the increase in quantity of trans-polyacetylene and the intensity enhancement of sp2 carbon scattering. Through the visible Raman spectrum detection, it is found that with the decrease of grain size, the intensity of 1140 cm-1 peak and 1460 cm-1 peak unceasingly strengthens, the intensity of the diamond peak constantly lowers, and the full width at half maximum (FWHM) of diamond peak continually broadens. The content of sp3 carbon decreases in the film. When micron diamond replaces cemented carbide as substrate material, the FHWM of diamond peak has greater decrease, and its intensity increases significantly. Impurity components in the films reduce significantly.

To improve the computational efficiency of computer generated hologram (CGH) and achieve the real-time nature of CGH, a novel fast CGH algorithm of triangle mesh models is put forward. In the algorithm, the Fourier spectral information of the triangle mesh is expressed as an analytic function by using affine transform theory and a Fourier transform does not need doing for per triangle mesh of the object. This algorithm reduces the amount of computation and improves the computational efficiency. The experimental results show that the new algorithm is simple and fast, which also can obtain the reconstruction image well. It is an effective method to improve the speed of CGH.

Aluminum-lithium alloy is considered to be one of the promising light-weight materials for aerospace applications. In the experiment, T-joint of 2060-T8/2099-T83 aluminum-lithium alloy with thickness of 2 mm is welded by using two high brightness solid state lasers from the two sides of the stringer aimed to integral aircraft fuselage panels. The weld porosity, cracking and joint mechanical properties are investigated. The results show that macro hydrogen pores can be avoided only when the surface layers are removed at least 0.15 mm thick, while the hot cracking can be completely eliminated with filler wire of AlSi12 and an appropriate bevel design. The hoop tensile strength of laser-welded T-joints is among 385~415 MPa, 82.5 % of that of the 2060-T8 skin sheet. The bond is the weakest region of the joint.

Dressing experiment of bronze-bonded diamond grinding wheels using a nanosecond pulsed fiber laser is carried out. The systematic research starts with the mechanism of laser dressing, followed by the analysis of ablated crater impacted by multiple-pulse and the studies of relationship between pulsed laser parameters and ablation effect. The laser dressing experiments is carried out at the foundation of laser ablation and the surface topography after laser dressing on wheels are then observed and assessed by optical microscope and grinding experiment.

The fit up precision is one of the facts that restrict the applications of the laser welding of the vehicle body components in a butt joint. An experimental study on gap margin in laser butt welding of 1.2 mm thick high-strength automobile steel DP800 is performed with hot wire filler under various butt gaps. Sound weld beads are obtained by optimizing welding parameters when the butt gap is not more than 100% of the steel sheet thickness (1.2 mm). The chemical composition distributions, metallographic phase and mechanical property of the weld are analyzed after welding. The result shows that the butt gaps have little influence on the chemical composition distributions and metallographic phase of the weld; the tensile strength of the weld is higher than that of the bass metal when the butt gap is not more than 83% of the steel sheet thickness (1.0 mm), while it reaches 89.9% of the tensile strength of the base metal when the butt gap is 100% of the steel sheet thickness (1.2 mm). The requirement of high fit up precision of butt joint can be loosened dramatically using laser welding with hot wire.

The influences of Nb and Cr on the microstructure and properties of the Fe-matrix laser clad composite coating reinforced by in situ particles are studied of Ti, W, Nb and Cr with different contents and ratios adding into the FeCSiB cladding powder, and in situ particle reinforced Fe-matrix composite coating is produced by laser cladding. It reveals that clad layers with good forming and homogeneous microhardness distribution are produced by adding Nb and Cr other separately or simultaneously. The particle distribution can be influenced by the contents of Nb and Cr efficiently. The microhardness of the layers increases firstly and then decreases with the content of Nb increasing, and decreases with the content of Cr increasing. The impact wear resistance is related to the toughness of the material. The micron carbide particles will not improve the impact wear resistance of the particle reinforced composite coating.

The influence of the double-clad fiber inner-cladding and core axial deviation on fiber splicing is theoretically analyzed. Based on the analysis of the results , in the master oscillator power amplifier laser system, a reasonable method to splice the key points is proposed. The 1509 W amplified power in all-fiber laser system at 1080 nm is achieved with a 533 W seed laser and a 1.16 kW pump power. The slope efficiency of the amplifier is up to 87.3% with central wavelength of 1080 nm, and the beam quality of the output laser is M2=1.46.

According to the reaction mechanism in non-chain pulsed DF lasers, a dynamical model for this kind of laser is presented by using rate equations theory. The optimum gas ratio and reflectance of output mirror are obtained by numerical calculation with this model using the Runge-Kutta method. The experiment is done based on UV-preionized discharge method. The main output parameters of non-chain pulsed DF laser gained from experiment agree well with the theoretical calculation. The theoretical analysis and experimental results both show that the output performance of non-chain pulsed DF laser can be improved by optimizing the parameters of working gas pressure ratio and the reflectance of output mirror. The laser has the best output performance when the working gas ratio is 101 and the reflectance of output mirror is 30%. The single pulse energy of 4.95 J, pulse duration of 148.8 ns and peak power of 33.27 MW are achieved under above conditions.

We demonstrate the ultrafast graphene oxide mode-locked Er-doped fiber lasers. The lasers can operate in all anomalous dispersion regime and normal dispersion regime respectively through the cavity design. When the laser operates in the all anomalous dispersion regime, the traditional soliton pulse trains at repetition rate of 61 MHz are generated. The measured pulse width is 500 fs, spectral width is 5.7 nm, and the maximum output power is 6 mW, corresponding to single pulse energy of 0.1 nJ. In the normal dispersion regime, dissipative soliton pulse trains at repetition rate of 11.5 MHz are generated. The pulse width is 143 ps, spectral width is 2.4 nm, and the maximum output power is 21 mW, corresponding to single pulse energy of 1.8 nJ.

A mounting configuration is presented to solve the loss of the third harmonic generation (THG) efficiency due to the gravitational effects. The analysis result of finite element model shows that the loss of the THG efficiency is nearly absent due to the gravitational effects when the original surface profile of the KDP crystals and mounting configuration is an ideal plane. Meanwhile, numerical simulation calculations are performed to analyze the influence of fabrication errors on the THG efficiency. Furthermore, the machining requirements of the peripheral milling have been discussed. The loss of the THG efficiency is nearly same under the different departure angles. The mounting configuration with the random error is acceptable, but the convex error and concave error should be controlled below 5 μm to realize efficient third harmonic output.

A model of a thermal blooming gas in a closed horizontal tube is used to study the effects of gas pressure on laser propagation. It focuses on the variation of beam quality in far-field and the switch time of heat conduction and heat convection in gas. By numerical simulation, the distributions of gas temperature, additional phase and beam quality and drift of mass center in far-field are shown under different gas pressures and absorptions of laser energy. And it is found that the change of beam quality is decided by both of the distribution of gas temperature and the thermal-optical coefficient, and the switch time of beam quality accords with the one between the two thermal conduction mechanisms, but the switch time of drift of mass center delays. Reducing the pressure in the tube, the gas has a higher temperature, but more even distribution, and the thermal-optical coefficient of gas scales down, so the beam quality consequently improves.

Based on the all-fiber ring-cavity multi-pass pulse amplifier, pulses of Q-switched with a few hundred nanosecond duration and repetition rate of 10~50 kHz range are amplified. Although the number of pass is not controlled, some important phenomena such as characteristics of gain and time domain evaluation in multi-pass amplification are acquired, and the output frequency of pulse is changed by the ring-cavity multi-pass amplifier. A gain of 7 dB higher than that of single pass amplification is obtained. Experimental results demonstrate that signal gain and conversion efficiency can be greatly enhanced by using ring cavity multi-pass amplification for a pulse repetition rate of 10 kHz or lower than that. The results can be helpful to investigation on all-fiber regenerative amplifier.

The symmetric dark core with a high intensity ring resulting from tight focusing of an azimuthally polarized beam plays a significant role in the fields of particle trapping and high-resolution imaging, etc. However, even in the well-corrected tight focusing optical systems, some aberrations inevitably occur. Numerical calculation is performed to analyze the influence of primary aberrations on the tight focusing properties of azimuthally polarized beams based on Richards-Wolf vector diffraction theory. The results show that the primary spherical aberration and the field curvature do not destroy the circular symmetry of the intensity distribution and a sharp high intensity ring can be obtained on the real focal plane. A positional displacement of the intensity ring takes place in the presence of primary distortion. The intensity ring is elongated in the presence of primary astigmatism. The primary coma tends to drive the intensity ring to move favorably on one side of the dark core, and the intensity null vanishes when the aberration coefficient becomes large.

A high efficiency 1064 nm Nd:YVO4 laser which is in-band pumped from high Stark level of ground state to upper laser level by another Nd:YVO4 laser operating at 914 nm is demonstrated. To overcome the drawback of low optical-optical conversion efficiency induced by poor pump absorption, the influences of doping concentration, temperature and length of laser gain medium on pump absorption and laser conversion efficiency are investigated in detail. On this basis, a high optical-optical efficiency of 56.9% is achieved by using Nd:YVO4 crystal with length of 20 mm as gain medium. The optical-optical conversion efficiency of the Nd:YVO4 laser in-band pumped from high Stark level of ground state at 914 nm reaches a practical level comparable to that of traditional pumping.

The problems of the large beam divergence angle and short detection range in the launch system of laser proximity fuze are investigated. By analyzing the optical structural characteristics of large field laser proximity fuze, combined with characteristics of distance influencing factors, and according to aspheric optical design theory, multilevel aspheric collimating lens are designed using Zemax software. Simulation results show that, comparing with single aspheric collimating lens, multilevel lens increase the brightness of the center of irradiation nearly 10-fold, and the spot size decreases and energy concentrates. At last, aspheric lens is processed and experiment has been done, the results show that this collimating system can compress spot size effectively, and the system detection range is improved.

A theoretical model of high power multimode fiber oscillator is built based on rate equations of the multimode Yb3+ doped fiber considering improvement of the mode coupling in the core. The effect of Yb3+ dopant profiles in the core on the output characteristics of fiber oscillator is carried out. The output characteristics of the high power multimode fiber oscillator with flat and parabolic dopant profiles fibers are described respectively. The output characteristics of the fiber oscillator with 75% flat dopant profiles fibers in different coupling coefficients are described. The results show that the different Yb3+ dopant profiles may have different impacts on the output characteristics. Specially, the fundamental mode output can be achieved by designing and controlling the Yb3+ dopant profiles.

The produced length of whispering-gallery-mode (WGM) lasing based on polydimethylsiloxane (PDMS) is investigated. It is found that the WGM lasing has a long lasing produced length pumped by evanescent wave along the optical fiber axis. A PDMS substrate with three fiber channels is fabricated. After injecting ethanol gain solutions of rhodamine 640, rhodamine B and rhodamine 6G separately into the three channels, three-wavelength WGM lasing emission at red (628～634 nm), orange (590～597 nm) and yellow (564～572 nm) bands is successfully observed.

The Haus master equation which describes the laser dynamics is solved numerically so as to study the effects of cavity parameters and characteristics of graphene saturable absorber on the pulses characteristics. Adopting the fast saturable absorber model, the relations between pulse characteristics with cavity length, graphene layers, transmittances of output mirror and beam radius on the laser medium are analyzed. In addition, the results are compared with that of semiconductor saturable absorption miror (SESAM). The conclusions show that the graphene and SESAM have different advantages. For example, for higher pulse energy, graphene is favorable for shorter cavity length and smaller radius on saturable absorber, while the SESAM is better for the opposite situation. The saturable absorbers and cavity parameters should be chosen properly according to concrete situations.

A high-power all-fiber Yb-doped picosecond fiber laser is achieved via a four-stage master oscillator power amplifier (MOPA) configuration. The seed source is a passively mode-locked fiber laser based on nonlinear polarization rotation (NPR) effect, with a central wavelength of 1062.8 nm, a repetition rate of 17.51 MHz, a spectral linewidth of 5 nm and an average power of 7.14 mW. To suppress the nonlinear effects in the process of power scaling, the repetition rate is upgrade to about 281.7 MHz through an all-fiber repetition rate extender. A 4.8 m large mode area (LMA) Yb-doped double clad fiber is utilized as gain medium in the main power amplifier. The maximal average output power of 31.2 W is obtained at the pump power of 60 W, corresponding to an optical-to-optical conversion efficiency of 52%. The output laser pulse has a central wavelength of 1063.7 nm, a pulse duration of 10.2 ps, a repetition rate of 281.7 MHz and a spectral linewidth of 7 nm. Moreover, the characteristics of the laser pulse are analyzed both in time and frequency domain.

High repetition frequency and low threshold laser with wavelength of 2 μm is reported using periodically poled MgO-doped lithium niobate (PPMgLN) with doped concentration of 5%. A 2 μm laser is obtained with PPMgLN pumped by a 1.064 μm laser and the corresponding relation between polarization period and the temperature is calculated. In experiments, the comparison of the efficiencies of two different polarization periods (32.2 μm and 32 μm) is carried on. We focus on the polarization period that has higher efficiency (32 μm) and conduct the experimental analysis of the oscillation threshold. Under the condition of the conversion efficiency is not less than 40%, the laser output with oscillation threshold of 0.046 MW/cm2 at near degenerate point(signal wavelength of 2.05 μm, idler wavelength of 2.236 μm) is achieved.

The spectral beam combining with an array of high-power and high-efficiency diode lasers is reported by using a common external cavity which contains a transmission grating, while simultaneously forces each element to operate at different, but controlled wavelengths and forces the beams from all the elements to overlap and propagate in the same direction. Using the diode laser array with front-mirror surfaces coating with high transmittance films, a continuous-wave laser with output power of 52.2 W and electro-optical conversion efficiency of 47.9% is achieved. The spectral linewidth is about 15.17 nm. Beam parameter product (BBP) is 3.7 mm·mrad, which is the same as that of single element of the array.

Defects in thin films are one of the most important factors influencing laser induced damage in the thin film components at ArF laser system. In order to analysis the defects of fluoride thin films in wavelength of deep ultraviolet (DUV), LaF3 and MgF2 single layers, LaF3/MgF2 half high reflection (HR) coatings and HR coatings are prepared by thermal evaporation on CaF2, the laser induced damage thresholds (LIDT) at the wavelength of 193 nm for the LaF3 single layer and HR coatings are measured. Nomarski microscope is used to observe the surface defects and damage morphology. It is found that the LaF3 single layer surface becomes more smooth after the preparation, while the MgF2 layer appears a few defects, and the defects increase as the layer number goes up. Based on micrograph of the thin films, lots of defects play a leading role in HR coating damage mechanism at this condition.

The HfO2/SiO2 films are deposited on K9 glass and Y3Al5O12 (YAG) crystal substrates by electron beam technology respectively. Nano-scratch tests are taken to investigate the mechanical properties of films respectively. The results show that the modulus of the films deposited on K9 and YAG are 34.8 GPa and 38.5 GPa respectively and the substrates have few effect on the elasticity modulus of the films. The adhesive force of the film is 7 mN on K9 substrate and 5 mN on YAG, and they present different failure modes. This can be attributed to the weak adhesion and large divergence of modulus between film and YAG crystal. The chemical binding state and elasticity modulus between the film and the substrate are taken to explain the different mechanical behaviors of the films on YAG and K9 substrates.

In order to enhance conversion efficiency and stability of thin film amorphous silicon solar cells, the nanosecond pulse laser is used to make the transparent conducting films textured. Transparent conducting film′s electrical, optical and crystal structure characteristics are discussed with changing laser parameters, such as laser power density, frequency repetition, etching speed and filling spacing. A comparison is made with the performance of cells manufactured by different texturing ways. The experiment results indicate that when laser power density is 0.85×105 W/cm2, etching speed is 600 mm/s, repetition frequency is 50 kHz, and filling spacing is 0.012 mm, the textured structure of transparent conducting films has less square resistance and more effective light trapping. It effectively improves the absorption, and enhances the conversion efficiency of cells.

In order to screen high reliability and stability area CCD from industrial class devices for the aerospace application, a board level screening and test method for area CCD is put forward and a corresponding screening and test system is designed. To find out unqualified devices, four steps are carried out, consisting of parameters test before aging process, photosensitive property test during the aging process, parameters test afer aging process and long-life experiment for the same batch sampling devices; the judgment of area CCD parameters is made via testing the camera which contains the area CCD; the integrating sphere and guideway are used to test the the bright field and dark field parameters in the darkroom environment. The integrating sphere, the target, the collimator and the six degrees of freedom adjusting system are used to modulation transfer function (MTF) maximum value test. The screening and test of area CCD have been used in engineering practice and passed through a series of test evaluation. The test evaluation results show that the screening and test meet the aerospace applications requirements.

To measure the sub-pixel image motion caused by satellite attitude instability or various disturbance, an effective approach to measure image motion by dual-CCD imaging system based on joint transform correlator (JTC) is presented. An auxiliary high-speed CCD is used to capture image sequences which are the same scenes as those images captured by the prime CCD. These image sequences are optically calculated by JTC system, and then relatively motion displacement can be obtained. The principle of this approach is represented, and measurement precision is analyzed in the condition of noise and image motion. Motion measurement system is also presented, and the experimental results show that the measurement precision can be controlled within 0.1 pixel.

To reduce the error in the measurement of object morphology, a method based on Fourier transform profilometry with optical fiber projection grating is proposed. The method analyzes the light intensity distribution that is projected on the receiver screen in theory. Its intensity meets certain Gaussian distribution by analyzing. The intensity correction is made in the process of fringe pretreatment and after correction the fundamental frequency components of stripes spectrum are clearer; the error that is brought by the non-sinusoidal stripe is compensated in the measurement of outcome. The simulation result shows that the accuracy of shape restoration has been greatly improved by the proposed method. The experiments on a standard object verify this method.

A Fabry-Perot (FP) resonant technique is proposed to measure the quadratic electro-optic coefficients γ33 and γ13 of transparent lead magnesium niobate titanate (PMNT) ceramic for transverse electric (TE) and transverse magnetic (TM) modes. The resonant wavelength in the PMNT FP resonant cavity will shift with the increase of applied voltage due to the quadratic electro-optic effect. The quadratic electro-optic coefficients γ33 and γ13 for TE and TM modes are measured, and their values are about 21.06×10-16 m2/V2 and -1.76×10-16m2/V2, respectively. Using the characteristic, a polarization-dependent PMNT electro-optic diffraction grating is designed and the diffraction patterns for different polarization light are also analyzed and compared in detail.

A method for automatically measuring wheel geometric parameters by laser scanning is presented. The running surface of wheel is scanned by the line structured light vision sensors, and the geometric parameters including wheel diameter, flange thickness, tread wear and rim width are calculated after processing the point cloud. This is different from the traditional methods which measures the geometric parameters by comparing a measured profile to the standard profile of wheel tread.This method is not strict with the relative position between the line structured light and the wheel. The feasibility of this method is demonstrated by experimental results.

An on-line three-dimensional (3D) measurement method based on low modulation feature for pixel matching is proposed. Only a fixed sinusoidal fringe is projected on the measured object moving with a constant speed. The object′s movement can produce the corresponding phase shifting deformed patterns. When several deformed patterns are recorded at equivalent steps by CCD, pixel matching must be done for these patterns in which the points are not one-to-one corresponding because of the measured object′s movement. On the basis of cutting the modulation information of the object from deformed patterns, Otsu algorithm is used to do the binarization automatically. The area having low modulation is segmented to be regarded as a featured template to realize the pixel matching and several phase shifting deformed patterns in which the points are one-to-one corresponding are acquired. The equal-step phase shifting algorithm is used to reconstruct the surface information of the measured object. Simulations and experiments show feasibility and validity of this method.

An approach to measure differential mode delay of multimode fibers based on broadband random signal is proposed. An amplified spontaneous emission (ASE) signal is divided into two beams, with one serving as the probe light and the other as the reference light. By correlating the reference light and the probe light transmitted through the multimode fiber, the differential mode delay is obtained by the correlation peaks. Differential mode delay for multimode fibers with different lengths is experimentally measured, and the signal-to-noise ratio of measurement is analyzed. Results show that the differential mode delay coefficient is 0.61 ps/m for the restricted mode launch condition. When the peak sidelobe level is 3.16 dB, the probe signal intensity is only 5.3% of the noise intensity.

Dynamic or static strain sensing and signal demodulation are realized by using random non-uniform sampling of superimposed fiber Bragg gratings (FBGs). Superimposed FBGs demodulating system based on tunable filter and its realization of non-uniform sampling method are presented. The superimposed FBGs are designed to generate non-equidistant space of sensing pulse train in time domain during dynamic strain signals measurement. By combining centroid hardware finding algorithm with smooth filtering method, the demodulation speed and accuracy can be simultaneously kept. Random non-uniform sampling of superimposed FBGs can be realized by generating periodic random variable triangle wave driving voltage of tunable filter. By using superimposed FBGs, a 1.9 kHz dynamic strain is measured by generating an additive non-uniform random distributed 2 kHz optical sensing pulse train at a triangular wave periodically randomly changing scanning voltage with the mean scanning frequency of 500 Hz. The dynamic strain measurement bandwidth is eight times of single FBG interrogation system.

Traditional passive optical network (PON) systems are based on binary non-return-to-zero (NRZ) modulation format, which results in some disadvantages, such as the lowest system spectral efficiency, Raman crosstalk against radio-frequency (RF) video channels and large fiber dispersion penalty. In order to avoid the above problems, applying orthogonal frequency division multiplexing (OFDM) modulation format in PON systems could be a feasible technique. So a 10 Gb/s intensity modulation and direct detection (IM-DD) 16 quadrature amplitude modulation (16QAM) OFDM-PON downstream transmission over 100 km standard single-mode fiber (SSMF) is experimentally demonstrated. The recorded downstream receiving sensitivity is -26 dBm at the bit error rate (BER) level of 1×10-3 using avalanche photodiode (APD) direct detection. The dispersion penalty can be ignored at 1556 nm wavelength when the transmission distance is less than 100 km.

The measurement of radar cross section (RCS) of scale models at terahertz can decrease the cost and time of the experiments, so it attracts wide attentions. In real measurement, generally a Gaussian beam or a similar beam source is adopted，while in the RCS estimation, usually a plane wave is assumed as the incident beam. In order to study the error caused by the assumption, the RCS of a perfect conducting flat plate is estimated in two dimension when the incident beam is a Gaussian beam. In the estimation, the influence of 2.52 THz collimated laser on RCS is studied. The estimation results show that choosing beam width of 40 mm can help to keep the amplitude of the error around 0.3 dB when the width of the flat plate is 20 mm.

To actualize precise polishing of silicon modification layer on space silicon carbide mirror, combined type polishing technology, consisting of computer-controlled optical surfacing (CCOS), flexible chemical mechanical polishing (FCMP) and ion beam figuring (IBF), is presented. The polishing principles, polishing machines, polishing experiments and polishing functions of three type polishing technologies are introduced, respectively. To improve figure accuracy and reduce surface roughness of silicon modification layer in a way, computer-controlled optical surfacing technology is adopted. The surface roughness and surface finish of silicon modification layer are improved further by flexible chemical mechanical polishing technology. The high figure accuracy of silicon modification layer is achieved finally by ion beam figuring technology. The surface modification silicon carbide plane mirror for space camera is polished by the combined type polishing technology. After polished, root mean square values of the figure and roughness of the space silicon carbide mirror are 0.014λ (λ=0.6328 μm) and 0.71 nm, respectively.

The regions of an interest image are the parts of priority attention and have more significance. The traditional visual attention model describes the information of regions of interest using fixed-size circles and can′t accurately express the outline of the regions of interest. A new automatic detection algorithm of the regions of interest based on adaptive radius search (ARS) is proposed. The new algorithm extracts color, intensity and orientation features of the image to generate a multi-scale saliency map. The global saliency threshold is calculated, which can get the end condition of searching the focus of attention. The adaptive radius search mechanism based on saliency ratio is proposed in the description of regions of interest to acquire the accurate information of regions of interest. The experimental results show that the new algorithm not only can effectively improve the detection precision of regions of interest, but also is more suitable to the features of human visual system. It has important value for the automatic target recognition of regions of interest in the future.

Image signal-to-noise ratio (SNR) will be reduced with the range increasing because of the influence of backscattering noise when blue-green laser radar is detecting and imaging underwater target. In order to solve this problem, by analyzing the noise types for the underwater target image of the laser radar, an algorithm of image pre-processing is presented, which includes the noise suppression in frequency domain and the improved Butterworth high-pass filters. The pre-processing algorithm can realize the image restoration, and improve the image SNR. The experiments of detecting and imaging the mine in indoor pool are carried out with the streak tube imaging lidar. The algorithm is used for the intensity image of the 5 m and 15 m underwater mine. The results show that the pre-processing algorithm can effectively suppress the backscattering noise of water. The contrast of image is improved. The clear object contour is obtained.

A phase-modulation surface plasmon resonance(SPR) sensor based on Mach-Zehnder configuration is presented and studied. Refractive index information of sample is extracted by phase substraction between the respective interference results of p-polarized and s-polarized light. Compared with the traditional phase-modulation SPR sensors, this design reduces the influence of environment noise and therefore greatly improves resolution. The system can distinguish 0.01° phase difference at least, which is equivalent to 10-5 RIU (refractive index unit). This system is employed to detect the reaction of biological molecules, and a resolution as high as 160 moleculars (IgG) per square microns is obtained. This study provides a certain reference to phase-modulation SPR sensors in application of bio-medical research.

The range resolution of a streak tube imaging lidar (STIL) system is researched based on the parameter estimation theory. According to the theory of mathematics statistics, the Cramer-Rao low bound (CRLB) of range resolution estimate variance is analyzed. Streak tube lidar pulse model is researched, and the formulas of CRLB are given. The influences of range resolution of the STIL system are analyzed. The results show that higher signal-to-noise ratio values and faster streak tube sweep rate are both helpful in range separation estimation efforts. In addition, there is an optimal pulse-width that produces an optimal range resolution for a particular streak tube sweep rate.