Gratings with many types of groove profiles such as rectangles and triangles can be fabricated by wet etching silicon, during which a key step is to accurately align the grating line of the mask with the 〈111〉 orientation of silicon. Sidewall roughness can be visibly reduced by high-precision alignment, which is necessary for the fabrication of large grating structures. A fanned alignment pattern was designed and fabricated, which was then etched to accurately determine the 〈111〉 orientation of silicon. After determining the 〈111〉 orientation, a grating with a height of 48.3 μm, period of 5 μm, and area of 15 mm×15 mm was fabricated successfully by ultraviolet lithography and wet etching. The aspect ratio of the grating was approximately 20, and the roughness of the lines sidewalls was 0.404 nm. A rectangular grating with a large aspect ratio and a triangular-grooved grating were fabricated successfully with holographic lithography and wet etching. The rectangular grating was 4.8 μm tall with a period of 333 nm and an area of 50 mm×60 mm. Its aspect ratio was approximately 100 with a sidewall roughness of 0.267 nm. The period of the V-groove grating was 2.5 μm, and its sidewall roughness was 0.406 nm.

In order to reduce the volume and weight of the underwater Raman system, and to improve its portability and detection sensitivity, a compact underwater Raman spectroscopy system with high sensitivity was developed and assessed. Through elaborate selection of components, a compact structural design was realized with both the weight and the volume well controlled. The size of the main body was kept at 795 mm in length and 260 mm in diameter, with a weight of 548 N, one third of the weight of reported DORISS (the first deep ocean Raman in-situ spectroscopy system). The laser was housed in the optical probe rather than in the main body, hence higher excitation efficiency was achieved with high power density. There are two advantages to put the laser head in the probe. A desirable excitation power density could be obtained without the consumption of laser beam during transmission in fiber, and better signal to noise ratio could be achieved without the stray light raised by the interaction of laser and optical fiber. In addition, 300 mW powered laser, efficient volume phase holographic grating and TEC cooled CCD detector were used to improve the system performance. The experimental results show that the LOD (limit of detection) of SO2-4 was less than 0.4 mmol·L-1. Its about four times than the value achieved by DORISS. Meanwhile the system can be used to identify minerals. All above prove the system to be highly potential in ocean exploration.

The diffuse attenuation coefficient of downwelling irradiance Kd(490) is of great value for evaluating the coastal water quality and estimating the optical depth accurately. In this paper, the performance of MODIS-Aqua Kd(490) products derived using the algorithms of Muller, Werdell, Morel, Lee, and KD2M with MODIS-Aqua remote sensing reflectance Rrs(λ) and Chl-a products were compared in the South China Sea (SCS) based on the relaxed match-ups between MODIS-Aqua products and in situ data collected from 2004 to 2012. Further, taking into account the unique optical properties of the SCS waters, the performance of Kd(490) products derived using the above algorithms with field reflectance Rrs(λ) and Chl-a data were evaluated in order to analyze the causes of the uncertainties of Kd(490) products and the necessity of regional algorithm optimization. The results show that MODIS-Aqua Kd(490) products derived using the algorithms of Muller, Werdell, and Morel with MODIS-Aqua Rrs(λ) and Chl-a products underestimated the in situ data, while the algorithms of Lee and KD2M overestimated the data, with average relative errors (APD) of 20.12%, 29.22%, 17.98%, 22.61%, and 21.61%, respectively. The algorithm of Morel performed the best, with a coefficient of determination R2 of 0.5. There are significant improvements in the accuracy of the above algorithms by using field Rrs(λ) and Chl-a data, especially for the algorithms of Werdell and Lee, with an APD of 5%—11% improvement. The algorithm of Morel performed the best, with an APD of 15.44%. It is necessary to choose the proper algorithm and conduct algorithm regionalization correction for the accuracy estimation of Kd(490) products in the SCS.

In order to realize supercontinuum (SC) generation with broadband spectral range covering more visible wavelength, a pump wavelength tunable all-fiber SC fiber laser was experimentally investigated. A tunable nonlinear polarization rotation mode-locked fiber laser was proposed, which could output picosecond dissipative solitons pulse with nine different central wavelengths from 1 041 to 1 076 nm. After two-stage power amplifier, the tunable laser with relatively high power was coupled into 10 m long photonic crystal fiber, SC with nine different output spectra was realized when the pump laser power was 500 mW. And it was found that when the pump laser central wavelength was 1 050 nm, the generated SC was with broad and flat spectrum and containing a large part of visible wavelength components. To further broaden the SC spectral range and obtain better flatness, the pump laser power was enhanced up to 1.45 W, the generated SC output power was 600 mW, while the short-wavelength boundary was 470 nm, 10 dB spectral bandwidth was 1 053 nm within 600—1 700 nm wavelength range.

To study a fiber grating sensing and demodulation system applied to satellites, ultra-short fiber Bragg gratings with grating length of less than 0.5 mm, reflectivity of over 40%, 3 dB bandwidth of over 5 nm, and the edge linear region of the reflection spectrum of over 2 nm were fabricated by the hydrogen enrichment of high germanium fiber and optimization of UV exposure power. A novel method was proposed to transmit the ultra-short FBG as a sensor and demodulate it by using its spectral linear region in this paper. The frequency stabilized laser whose central wavelength was located in the spectral linear region was incident on the ultra-short fiber grating. The reflected optical power changed with the drift of the ultra-short fiber grating spectrum. As the frequency stabilization laser was located in the linear region, the reflected optical power was linearly related to the spectral drift, so that the sensing measurement could be realized. Using the system for strain and temperature measurements, the experimental results show that the optical power has a good linear relationship with strain and temperature, and the linearity is 0.997 and 0.999, the sensitivity is 54.59 nW/με and 230 nW/℃, respectively. The method can be used for accurate measurement of temperature and strain, and has the potential advantages of simple structure, low power consumption and high measurement spatial resolution.

To facilitate zoom lens design, a systematic zoom lens design approach, based on mixed lens modules, is proposed in this paper, which consists of paraxial lens design, lens type determination, and aberration design. During paraxial lens design, Gaussian brackets and matrix optics are adopted, with which the relationship between component interval distances and component focal lengths can be easily determined. A lens module classification model is trained, with which the lens type can be determined from its thin lens parameters. More than 80% of the lens components have been successfully classified by the classification model. Mixed lens modules, namely thin lens modules and thick lens modules, are applied for aberration design of lens components, with which the component aberrations can be easily balanced and some practical considerations, such as interval distances between lens elements and components collisions, can be handled simultaneously. At last, an example is provided, which shows the detailed procedure of zoom lens design using mixed lens modules.

In order to meet the demands of a mobile vehicle-based adaptive optical system, the traditional wave-front processor hardware architecture was redesigned. In this study, we combined a wave-front processing board, wave-front processing daughter board and DA conversion board, using optical fiber for board communication. In this way, the reliability of the system was improved while the functional requirements were satisfied. Meanwhile, the wave-front processor is the computing center of the closed-loop control of the adaptive optical system, and its operation delay directly affects the control bandwidth of the system. This paper presents a multi-pipeline adaptive real-time wave-front processing method based on FPGA to complete the slope calculation, restoration calculation, and control calculation. For the adaptive optical system, which can provide two-stage precision tracking and consists of 97 sub-apertures and 97 deformation mirrors, the processing delay is 506.25 μs, meeting the requirement of 1 500 Hz real-time wave-front processing.

In order to research the applied advantages of composite crystals to the passively Q-switched lasers, a compact diode-end-pumped passively Q-switched laser was built up with a Yb∶YAG/Cr4+∶YAG/YAG composite crystal for generating efficient 1 030 nm and 515 nm pulsed lasers. The effects of pump power and initial transmission T0 on the laser properties were studied. As a result, a 1 030 nm laser with an average power of 1.97 W is obtained at T0 of 95% with slope efficiency of 33%, and the pulse peak power reaches to 87 kW at T0 of 85% with pulse width of 3.14 ns. In addition, in the frequency-doubled experiment with LBO, a green laser with higher output power of 634 mW is obtained at T0 of 90% with slope efficiency of 11.2% and pulse width of 15 ns. Finally, the red shift of spectrum and the leakage of output mirror were studied, as well as the method for increasing output power of green laser.

The existing Digital Light Projection (DLP) systems, in which the components such as color wheel and relay lens are used, resulting in a complex projection system. Aiming at this defection, a new DLP projection system was designed, where the three-color LED were used as the illumination light source and refractive index lens were used to produce the parallel light. Compared with the conventional projection system, components such as the color wheel, the relay lens, the mirror could be saved. Besides, the structure of the optical system was simplified. The simulation results and ray tracing by the Tracepro software show that the light spot uniformity reaches 96.9%. The results indicate that optical path of the projection system is simplified, and uniformity of the spot is also improved. The design only need to select the appropriate projection lens, that it can eliminate the influence of extended LED on the projection spot.

To measure the residual stresses in thin films with limited area prepared by pulsed laser deposition, as well as to solve the problem of the relatively large error introduced by the Stoney equation under certain circumstances, a measurement method for residual stresses in thin films based on cantilever structures and numerical calculation was proposed in this article. In this method, atomic force microscopy probes with zero initial curvature were used as substrate cantilevers, and thin films were deposited on the substrate cantilevers using pulsed laser deposition. The bending profiles of the substrate cantilevers before and after thin film deposition were recorded and used in numerical calculation, together with other parameters including the thin film thicknesses, the geometries of the substrate cantilevers, and the Young’s moduli and the Poisson’s ratios of the involved materials, to analyze the experimental data and obtain the residual stresses in the thin films. By using this method, the residual stress in vanadium dioxide thin films, prepared by pulsed laser deposition in a high-temperature environment, is measured to be -340 MPa, corresponding to the value reported in the literature. The measurement method for residual stresses in thin films proposed in this article, which is based on cantilever structures and numerical calculation, has the advantages of wide versatility, good accuracy, and low costs.

Modeling compensation of nonlinear friction is vital to improve the trajectory tracking performance of high acceleration motion systems. To overcome the problem of inaccurate estimation of the start-stop stage nonlinear friction (including friction overshoot) associated with the traditional parametric model for high acceleration motion systems, this paper describes a novel extended parametric model combining the traditional model structure with the extended Stribeck model. The training data for identifying the model parameters are obtained using the high-precise Iterative Learning Control (ILC) approach, which supplies the nonlinear friction feed-forward compensation data with limited trajectories in the workspace. The data are fitted with the Levenberg-Marquardt algorithm. Finally, the proposed model is validated with different trajectories on a high acceleration position platform driven by a Voice Coil Motor (VCM). The experimental results indicate that the proposed method can overcome the influence of nonlinear friction associated with the traditional model, including the friction overshoot in the start-stop stage. Moreover, the accuracy is comparable with the result of ILC, but offers the advantage that the proposed model can avoid the problem of poor generalization in ILC to realize the friction compensation of an arbitrary trajectory in the workspace.

In order to improve the rapid positioning of precision mechanical devices and the amplification ability of displacement amplification structures for the output displacement of piezoelectric stacks in piezoelectric linear motors, a new kind of double-foot driving piezoelectric linear motor with secondary lever and flexible hinge was proposed. First, the operating principle of the motor was analyzed and the trajectory equation of the driving feet was formulated. Next, the parameters of the straight circular flexure hinge were optimized in order to improve the output performance; the optimal hinge parameters obtained were thickness th=0.2 mm, cutting radius Rh=1 mm, and width bh=10 mm. Finally, the prototype motor was designed, and the amplitude, speed, and load performance were tested. The motor speed was analyzed based on the orthogonal test method; our analysis revealed the sensitivity of the motor speed on the voltage. Our experimental results revealed displacement amplitudes of the driving feet Ⅰ and Ⅱ in the range of 75 μm and 63 μm, respectively, i.e., the difference is about 12 μm. Furthermore, the speed was observed to reach 16.163 mm/s under a voltage of 120 V and a signal excitation of frequency 110 Hz, and a maximum load capacity of 1.7 N was obtained. Compared with existing piezoelectric linear motors, our motor possesses the advantages of a simple structure, easy installation and debugging, fast response, large amplitude, high speed, and stable operation.

To improve the trajectory tracking performance of a spatial rigid-flexible 3-RRRU parallel manipulator, a nonlinear control strategy based on a multibody inverse dynamic solution by means of a transient kinematic correction method was proposed. First, the nonlinear inverse dynamics of a spatial 3-RRRU parallel robot with flexible links was developed according to both the Natural Coordinate Formulation (NCF) and the Absolute Nodal Coordinate Formulation (ANCF). The derived models consider the shear deformation and can describe the large deformation for each beam. By analyzing the compatibility problem during the solution process of the rigid-flexible dynamics of the close-chain mechanism, we were able to develop the transient kinematic correction method and the derived stable causal solutions according to the NCF and the ideal kinematic model. Finally, the control strategy for the manipulator is presented, which was based on the solutions and simulations, and experiments were performed to verify the feasibility and effectiveness of the method. The results showed that the solution precision of the inverse dynamics was 10-6 and that the compatibility error of the constraints was 10-8. Compared with those based on the control strategy of the rigid parallel mechanism, the maximum tracking error and the roundness error of a prescribed circular trajectory based on the provided control strategy can decrease by 0.465 mm and 0.416 mm, respectively. The presented method can solve the compatibility problem of multibody dynamics with constraints, thus effectively improving the overall convergent performance of a dynamic system. The control strategy can provide better tracking performance for the parallel mechanism.

In this study, a novel type of lens shutter with compact structure, reasonable layout and high reliability was designed for use in digital aerial photography mapping cameras. The double-blade shutter achieved the same shutter efficiency as a five-blade shutter. A coaxial gear train with different transmission ratios drives the double-blade rotation by miniaturized design, thereby controlling exposure. The speed of the blade is varied continuously by controlling motor speed, which helps attain the appropriate exposure. Simulation and experimental analysis of the shutter indicates that the range of shutter exposure time reaches 1/100 s-1/1000 s, and the shutter efficiency can increase to 85%. The shutter with wide exposure time and high shutter efficiency is found to have small effects on optical distortion, and sufficient stiffness to meet the camera requirements.

There are strict requirements for the deformation of glass plate in the vertical direction in the process of on-line detection of LCD glass substrate. In order to determine the nozzle requirements in index spacing, the theoretical analysis of the LCD glass plate supported by gas suspension was carried out based on the small deflection bending theory of elastic sheet, and the LCD glass sheet maximum deflection formula under the action of the blowing and suction nozzles was derived. The analysis shows that the maximum deflection is proportional to the square of through-hole spacing and degree of load; The load distribution and the maximum displacement of the glass substrate under different nozzle through-hole spacings were obtained by the Fluent simulation on the LCD glass substrate gas suspension system. The maximum displacement of the glass plate, pressure distribution of the gas film surface and economic aspects were considered to obtain the range of the through-hole lateral spacing of the gas film unit. A type of nozzle layout parameter of the optical detection apparatus for the LCD glass substrate using gas suspension transmission was presented; the reasonable nozzle spacing should be between 20 mm and 25 mm.

In consideration of the issues of small output force and low speed of inchworm linear actuator, a novel inchworm linear micro actuator based on compliant mechanism was presented. The micro actuator was composed of two clamping mechanisms, a driving mechanism and an output shaft. The movement was characterized by the drive mechanism, which drives the clamping mechanism to create reciprocating linear motion, and the clamping mechanism, which was driven by linear motion from the output shaft. The flexible lever structure was adopted to design both the clamping mechanism and the driving mechanism in order to offer enough clamping force and driving force and improve the speed of the micro actuator. The relationship between the driving voltage and clamping force of the clamping mechanism, and the relationship between input displacement and output displacement of the driving mechanism were established by the pseudo rigid body method. The relationship between the input force and the driving force was established based on the function principle. The prototype of micro actuator was made and an experimental platform was set up to test the performance of the micro actuator. The experimental results show that, when the driving voltage is 120 V, frequency is 95 Hz, and maximum speed is 48.91 mm/s; the maximum clamping and driving forces are 216.43 N and 13.5 N.

In order to realize intelligent grinding and machining of workpieces with complex surfaces, robotic grinding process planning for complex blade surfaces was performed.The position planning algorithm of the grind point and the posture planning algorithm based on the maximum contact principle were studied. First, the grind path was obtained through the secant transverse line cutting method and described by NURBS curve, and then the curve feature parameters were extracted and grinding position planning was performed according to the set threshold. Then, based on the maximum contact principle of the grinding wheel andworkpiece, the posture planning of the grind point was presented. After that, the complete grinding path was obtained.Then, the position and posture data were converted from workpiece coordinate system to TCP coordinate system. Finally, the simulation platform of flexible grinding system was constructed to generate robot control program. Experimental results indicate that the proposed path can be used for robotic grinding of blade complex surface. The blades are grinded by using the path obtained by proposed planning method and the path of CAM software planning, and the corresponding surface roughness is 0.695-0.930 μm and 2.803-3.243 μm respectively. Therefore, the proposed method can be applied to the grinding path planning of complex surface. It ensures that the tool and the workpiece are in maximum contact, and thereby avoids uneven grinding caused by the poor position and pose.

To obtain a frequency-selective surface (FSS) structure with band-pass and band-stop spatial filtering functions at low frequency in different time domains, a design method for switching between band-pass and band-stop responses of the FSS was proposed. It was consisted of mounting PIN diodes on the miniaturized FSS, which was designed based on a convolution technique, and utilizing EM/Circuit Co-Simulation for calculations and analysis. It was noteworthy that the convoluted pattern was not only the filtering structure, but also serves as the feeder. When PIN diodes were in the ON-state, the parallel LC circuit constituted by the inductance L1 of the metallic patches and the capacitance C1 of the pattern slots represents the band-pass performance of the FSS. Conversely, the total capacitance C2 of the reversed PIN diodes and the gap between the pads in series with the inductance L2 of the metallic patches indicate the occurrence of band-stop performance for the FSS when the PIN diodes are in the OFF-state. Using a printed circuit board and surface mounting technology, a prototype of 400 mm×400 mm was fabricated and measured by the free space method. The simulation and test results indicate that, for a frequency of 2.45 GHz, the FSS exhibits strong transmission when the PIN diodes are in the ON-state, whereas a strong reflection is obtained when the PIN diodes are in the OFF-state. With this method, arbitrary switching of the FSS between band-pass and band-stop responses is realized, based on electrically controlled PIN diodes, which has a broad range of prospective applications in the fields of telecommunication, electromagnetic shielding, and radar stealth.

In order to realize quick response and effective detection in emergencies such as flight missing and natural disaster, an optimize-by-priority on-orbit task real-time planning algorithm based on priorities of targets was proposed. First, ground targets were divided into several priorities in order of importance, and the task planning was transformed to an optimization problem with objective of maximizing the imaging gain by analyzing constraints between target locations and satellite performance. Then, the targets of the same priority were selected by the rule of “first come, first served”, lower-priority targets were inserted into current optimal imaging sequences filled with several former-priority targets, and the maximum imaging gain will be obtained. Finally, a 500 s planning simulation of 18 ground targets was given. Results indicate that the algorithm proposed can give imaging sequence with the maximum gain within 1 s. The proposed algorithm calculates faster and gives more stable results as compared with the genetic algorithm.

To decompose asymmetric full-waveform LiDAR data with unknown number of components, a full-waveform LiDAR decomposition method was proposed based on skew-normal distribution and reversible-jump Markov Chain Monte Carlo (RJMCMC) algorithm, which can automatically determine the numbers of components. First, the energy function was used to describe the differences between the actual waveform and the ideal waveform that obeyed the skew-normal distribution, and the likelihood function was defined by Gibbs distribution. Second, the parameter models of the ideal waveform were established using the prior distribution. Then the Bayesian paradigm was followed to build the ideal waveform model. Third, an RJMCMC algorithm was designed to determine the numbers of components and decompose the waveform. The proposed algorithm was used to decompose ICESat-GLAS waveform data in various typical regions. Experimental results indicate that the cross-correlation of the true data and the result is up to 98.9%. The proposed method can not only fit the skewed waveform data and normal waveform data, but also more accurately determine the number of components in comparison to other methods. It can realize the accurate decomposition of full-waveform LiDAR data, and the decomposition result is consistent with the corresponding elevation information.

An efficient method for extraction of planar features from point clouds was proposed based on the concepts of self-adaptive octree voxel generation and voxel-based region growing. The proposed method involved the selection of correlated thresholds through statistics of voxel information. A voxel-based region growing approach was employed for planar feature extraction, instead of a point-based one. A point cloud was voxelized in initial voxel width and the geometrical features for each voxel were calculated, including the normal vector, eigenvalue, and three dimensionality features. The terminal constraints for octree subdivision were thereby determined through statistics and a list of octree voxels with inhomogeneous sizes was obtained after subdivision. Furthermore, planar facets were extracted through voxel-based region growing at different levels associated with the corresponding statistical threshold constraints. Evaluation experiments were performed by analyzing four different types of point clouds. The obtained experimental results show that the precision and recall rates can reach 95%, which indicates that the proposed method is insensitive to data quality and can be adaptive to various laser-scanned point cloud data. The proposed method can therefore achieve fine planar feature extraction results with high operating efficiency.

Stereo reconstruction from image pairs is a genetic method for 3D acquisition of human faces. Depending on available imagery and reconstruction algorithm, the resulting 3D reconstructions may have deficits. We improve generic morphable model and fused with stereo reconstruction to remedy such deficits. Firstly, we obtain the face bounding box by Max-Margin Object Detection, and recognition the face feature points directly from a sparse subset of pixel intensities by the Ensemble of Regression Treesmethod. Secondly, through the color PCA model to generate the shape and color of the 3D facial statistical model, the ISOMAP algorithm is used to convert the 3D mesh to 2D surface and extract the texture information to get the facial model.Finally, the surface registration with two step non-rigid change on source mesh: we mean the global deformation in the source by using few anchor points on the source mesh, and using the Radial Basis Function (RBF) to express non-rigid global deformation; and then Procrustes analysis is applied for non-rigid transformation of the source vertex, the k-nearest neighbor transformation through weighted scheme tosmooth the local deformation. The experiment result shows the high-quality scan which is used for cloud comparison with the face model, stereo reconstruction and our deformed face model, after aligning all models with the high quality scan, three RMS facial deformation model values are 2.795 2, 2.102 8, 2.153 4.Comparing to other models, deformed face model has smaller discrepancies dominate more strongly, and the deformed face model is visually more shape consistent to the image. Qualitative and quantitative analysis of the deformed face model shows that the combination of stereo reconstruction with general shape information about human faces is geometrically superior to both the reconstruction based on single image from generic model as well as the stereo reconstruction without consideration of the generic model.

To improve the performance of the clutter suppression and small target detection, a detection algorithm for a small target in sea clutter was proposed, based on the spatio-temporal chaos analysis. First, the sea clutter phase space was reconstructed as a chaotic dynamical system, and the chaotic parameters of the sea clutter sequence image were extracted to verify that the sea clutter owns chaotic properties in the spatial and temporal domains. Furthermore, the spatial chaotic reconstruction function, the temporal chaotic reconstruction function, and the space-time coupling coefficient were estimated by the radial basis function neural network. Finally, the spatial and temporal chaotic functions were integrated jointly to reconstruct the spreading regularity of the moving sea clutter. Some experiments were induced on the small target in various fluctuating sea clutter, and the experimental results show that the proposed algorithm, can improve the performance of the sea clutter suppression and enhance the small target detection ability, with the prediction error reduced by 10% and the detection probability increased by 20%.

A method for remote sensing image retrieval based on convolutional neural networks was proposed. First, the convolution and pooling of remote sensing images were conducted by multi-layer convolutional neural networks. The feature maps of each image were obtained, and the high-level features were extracted to build the image feature database. In this process, the training of networks parameters and the Softmax classifier were completed using feature maps. Then, in the image retrieval stage, classification was introduced by the softmax classifier which will improve the accuracy of image retrieval. Lastly, the remote sensing image retrieval was sorted based on the similarity between the query image and database. Retrieval experiments were performed on the high-resolution optical remote sensing images. The average retrieval precision on five kinds including water, plant, building, farmland and land is 98.4%, and the retrieval precision on seven types (adding plane and ship) is 959%. The introduction of class information improves the retrieval precision and speed, saving time by 17.6% approximately. The proposed method behaves better than the methods that based on color feature, texture feature and the bag of words model, and the results show that the high-level feature from deep convolutional neural networks can represent image content effectively. Experimeat indicates that retrieval speed and accuracy of optical remote-sensing images can be effectively increased in this method.

The relation between ground distortion and instantaneous field of view of the imaging system was analyzed; variable-speed scanning imaging was proposed and angular scanning speed formula was derived with infrared zoom scanning imaging system of super-swath low distortion in space as research object in this paper. Meanwhile, a kind of rapid fly-back method of sine acceleration was designed to solve the shortcomings such as low scanning efficiency of uniform 360° rotation, scanning line appliance needed to be installed for two-way swing scanning imaging and high complexity and low reliability of the system. Simulation and experiments were conducted on variable-speed scanning and rapid fly-back method of sine acceleration in this paper and the result indicates that the state shift between slow scanning and rapid fly-back of scanning control system is stable and the start and stop angle error of scanning is just 1.44 arc-second; scanning speed stability is ±0.5%; time error in scanning imaging process is 83 μs and fly-back time error is 250 μs; time deviation of the whole scanning period is less than pixel integral time (355 μs) and scanning efficiency reaches 86%, thus scanning efficiency is improved and impact and vibration of scanning mechanism are reduced meanwhile to satisfy imaging requirements. Rapid fly-back method of sine acceleration will provide a reference for rapid fly-back motion design of airborne infrared scanning imaging system.

Research on the lightning activities by taking advantage of the lighting detection data from satellite and ground-based observations is of great importance for lightning monitoring and early warning for thunderstorm. A satellite-based lightning imager can detect lightning signals over both ocean and land areas, and provide total lightning data including intracloud and cloud-to-ground lightning. However, a ground-based lighting location system usually gets the cloud-to-ground lightning. The two ways detect optical and radiometric emissions from lightning respectively. Firstly, the differences between satellite and ground-based lightning detection were analyzed in the aspects of the observed objective, location accuracy, and detection efficiency. Secondly, based on the data of TRMM LIS and the ground-based lightning detection network over China, a method for combination of satellite and ground-based lightning detection data was proposed. More specifically, by matching the similar units from both kinds of data and clustering the lightning groups, the combined lighting flashes could be extracted. Finally, the method was used in analyzing the lightning activities in southwest China from 2008 to 2014. The results show that the proposed method is feasible from the perspective of climatic and topography factors. In future, by integrating the data of geostationary satellite-based lightning imager, the matching error can be decreased, and the improved method will be applied to more solutions.

In order to ensure the high precision of dynamic visual measurement of rail abrasion, image acquisition and image processing technology were used to get clear light-stripe image and extract the sub-pixel coordinates of the light-stripe accurately. First, according to the high contrast between the light-stripe and the background brightness, a method of automatic exposure of the camera based on the brightness of the light-stripe was proposed, which could ensure a clear image of the light-stripe. Then the dynamic threshold segmentation method was presented to extract the light-stripe preliminarily according to the lightness decay feature of the light-stripe in the normal direction, this method can filter out the background of the image while preserving the normal attenuation of light. Finally, according to the overexposure information of the image to determine the approximate position of the center point of the light-stripe, then the Hessian matrix was calculated for the corresponding pixel position of the segmented image to obtain the sub-pixel coordinates of the light-stripe. Using MFC to write application for testing, under different illumination and background interference, this method can accurately extract the center of light-stripe, the extraction accuracy is 0.05 pixel and the computation time is reduced by 40% compared with the classical Steger algorithm. Experimental results show that the proposed method has good stability and strong anti-interference ability, can meet the requirements of the rail abrasion measurement.

In order to overcome the shortage of traditional optical passive ranging system such as low stability, difficult to guarantee insufficient ranging accuracy in long-distance condition and difficult to reduce the system volume and weight, a method for visualization of optical passive ranging system based on target feature size was presented in this paper. Passive ranging model was established and ranging principle was analyzed. Theoretical analysis was carried out to obtain the distance inversion formula and the ranging error accuracy. Through hardware and software design, the visual passive ranging function of the system was achieved and the all-weather image information acquisition was realized. Finally, target distance was retrieved by the target feature size and the imaging system parameters, real-time ranging was realized during the imaging process, and the passive ranging experiment was carried out. Experimental results show that the accuracy of passive ranging is better than 10%. Under the current hardware and software configuration, the target ranging distance is greater than 1km. The method has good robustness and stable performance. It can be widely used in the practical engineering of all-weather target imaging information acquisition and photoelectric passive ranging.