This PDF file contains the front matter associated with SPIE Proceedings Volume 11567, including the Title Page, Copyright information, and Table of Contents.

In recent years, terahertz imaging system has become more and more mature. There are good detectability for hidden objects on the human body surface, because terahertz waves has certain penetrating characteristics and low electromagnetic radiation in terahertz band. Using terahertz imaging technology has the advantages of high efficiency and non-radiation to carry out human body safety inspection, that makes it more and more widely used in the security security field. In order to achieve the purpose of detecting the hidden objects on the human body surface using terahertz technology, a convolution neural network named ResDeepNet is proposed in this paper. This network is different from the convolution neural networks structure of traditional optical images. According to the terahertz imaging characteristics of human body image, the semantic segmentation network is designed. Through the convolution network, the segmentation algorithm of terahertz image is realized, the target information is enhanced, and the noise is effectively suppressed in the terahertz image. The segmentation network and algorithm achieved the best results in test and practical application.

Asteroid exploration has become one of the important ways to understand the origin of the universe, exploit the space mineral resources and protect the earth from the asteroid impact. The imaging spectrometer, which can acquire the spatial and spectral information simultaneously, has become one of the most important payloads for the asteroid exploration. In this paper, the optical design of a broadband Offner imaging spectrometer for the asteroid exploration is presented. It covers from 0.4μm to 3μm, and the F/# is 6 and 3 for the VNIR band (0.4μm-1.0μm) and the SWIR band (1.0μm -3.0μm), respectively. The convex grating in the imaging spectrometer is divided into two regions of different groove spacing, and each region is optimized with different blazed angles to improve the spectral response of this imaging spectrometer. The diffraction efficiency is analyzed with the Comsol software and the result is greater than 0.36 in the wavelength range of 0.4μm to 3μm. After optimized with the optical design software, the diameter of the imaging spectrometer's spot diagram is less than one pixel, and the lowest MTF is 0.460 at the Nyquist frequency. The smile and keystone is less than 10% of the pixel.

Most of the existing defect detectors focus on the size, location, depth and number of defects of the tested components. The instrument is usually large in size and requires high accuracy for the environment and motion devices. In contrast, the direct random bed motion, which aims at finding and locating defects, is highly efficient, low-cost and environmentally practical, while the research on vibration-resistant defect tester is rare.In order to solve this problem, based on the principle of micro-scattering imaging in dark field, a set of optical component surface defect detection device is built, and the influence factors of light intensity, illumination angle, wavelength and other defect detection factors are experimentally studied, in order to provide design basis for the follow-up development of on-line defect detection instrument. The experimental results show that the most important factor affecting the sensitivity is the azimuth angle and pitch angle of the incident light, which is more than 30 degrees between the incident light and the scratch direction. When the pitch angle is between 60 degrees and 70 degrees, the higher detection sensitivity can be obtained. In addition, improving the illumination intensity can help to improve the detection of defects. In the visible range, the wavelength has little effect on the sensitivity.

In analogy to the velocity decomposition in continuum mechanics, we introduce two new tensors, referred to as the deformation-rate tensor and rotation-rate tensor, to the optical coherence theory for stochastic optical field, and decompose a coherence current vector into its translation, rotation and deformation components to study the optical coherence dynamics. To investigate the optical coherence propagation and evolution, we have conducted an experiment with results given to demonstrate the two newly introduced tensors.

Through the investigation and analysis of the characteristics of Martian atmosphere, the distribution of solar energy outside the Martian atmosphere is calculated. Using the fast calculation algorithm to calculate Mars surface radiation, a fast calculation mode is established, and a visual software interface is written. The radiation under given conditions can be calculated by directly inputting the location, time, solar zenith angle, surface albedo and light depth. The calculation results of the fast calculation mode are compared with the other two models, which increases the credibility of the calculation results. The calculation results are used to analyze the variation of surface radiation of Mars with solar zenith angle and light depth. This study will help researchers to predict the location and time of landing of Mars detectors, and provide theoretical support for analyzing the solar power generation situation of Mars. To ensure the successful implementation of the Mars exploration program.

In this paper, the finite element model of the fiber optic plates is established by using the finite element software. The simulation process is basically in line with the actual production process of the fiber optic plates. According to the simulation results, the deformation degree and speed of each part of the fiber optic plates in the process of melting pressure, as well as the changes of stress and strain of each part in the process of forming are analyzed. The results show that the deformation speed and degree of different parts are different in the process of melting pressure of fiber optic plates, especially the upper and lower end faces and side edges of fiber optic plates; and the stress and strain of each part are constantly changing, and the stress and strain values of the upper and lower end faces and side edges of fiber optic plates are larger than others.

Spinal surgery, such as pedicle screw placement, is difficult and risky due to the complexity of the physiological characteristics of the spine. Although intraoperative image guidance using a C-arm or an O-arm can significantly improve the success rate of the procedure, it is time consuming and the person is exposed to X-ray radiation for a long time. To solve this problem, this paper aims to exploring the application of new real-time 3D imaging technology in spinal surgery. Firstly, a CT scan of the pig's spine bone about 20 cm in length was performed and the accurate 3D data was obtained. Then, the 3D reconstruction was performed using a TOF camera and a light field camera. At last, the accuracy of the results was compared. The experimental results show that although the TOF camera has a good real-time performance, its resolution and accuracy of 3D reconstruction are low. The multi-frame superposition method can improve the data accuracy. The real-time performance of the light field camera is slightly lower. And the point cloud acquired by the bone is sparse because the spine texture feature is not obvious. But the accuracy of the 3D reconstruction can be significantly improved after the aid of structured light illumination. Therefore, the two techniques about 3D imaging proposed in this paper have some certain development potential in real-time navigation for spinal surgery.

The BaSO₄ diffuser and the PTFE diffuser are general reference diffuser in BRDF measurement. However, there are some differences of reflectance charactors between these two diffuser with illumination of narrow laser beam, and it affects the BRDF measurement precision. So the study of the differences of BRDF between the diffusers is necessary. Analysis the BRDF component of the diffusers was made with proposal of a 7-parameter model which including the volume scattering. Fittings of parameter models to the BaSO₄ diffuser and PTFE diffuser was made based on Simulated Annealing Algorithm respectively. The results showed that the 7-parameter model decreased the variance and improved the PTFE diffuser’s BRDF fitting results compared with 5-parameter model.

A bio-inspired Lidar imaging method with a non-uniform curved linear array receiving system is proposed. The receiving system is composed of a curved surface, a microlens array, an aperture array, and an avalanche-photodiode (APD) array. The microlens array and APD array are placed on a curved surface to mimic the large field of view (FOV) feature of the compound eye. The location distribution of the microlens array and APD array on the curved surface is non-uniform to mimic the retina-like property of the human eye. An experimental prototype, which consists of 12 microlenses and APD detectors is built and the effectiveness of the proposed approach has been validated by experiments. The results show that the whole FOV of the proposed system is 25.8° and it has a higher resolution in the central FOV and the lower resolution in the peripheral FOV. Moreover, the results show that the proposed Lidar imaging system has the rotation and scaling invariance property. This method is beneficial for developing a Lidar system with large FOV and high imaging efficiency.

Aiming at the problem of unclear results and low accuracy of ultra-high speed photographic measurement such as the transient process of glass crack breaking at the end of projectile, a laser source with high power and uniform facula was designed, and the results were verified by software simulation and field test. First, the specific scheme was designed according to the parameters of high-speed camera. High power laser beam was generated by semiconductor laser array, and the compound eye lens was used to divide and superimpose the shaping beam to form uniform rectangular facula. Then, the mathematical model of the facula and the energy model of the facula were established to analyze the homogenization mechanism of the large area rectangular facula. After that, Zemax software simulation proved that the size, uniformity and illumination of the facula could meet the needs of high-speed photography. Finally, the experimental verification proved that the facula could compensate the background light. The final results showed that the designed large area light source had 425 W power and 0.9203 facula uniformity. It basically satisfied the requirements of brightness and uniformity of background light source for high speed photographic detection.

An in-fibre temperature sensor based on a novel dual-mode fibre is proposed and experimentally characterized. The sensor head uses an in-fibre Mach-Zehnder(MZ) structure to sense the surrounding temperature, which depends on the interference spectrums. The concentric multilayer-core fibre (CMCF), in which only two modes could be propagated, would be easier to fabricate an in-fibre MZI with clean spectrum than many fibres, such as multimode fibres, thin-core fibres, PCFs and et al. And low-loss high-strength splice between CMCF and conventional single-mode fibre can be implemented with a commercial fusion splicer. Both the propagation characteristics and operation principle of such a sensor are demonstrated in detail. A sensitivity of ~50.13 pm/°C within temperature range of 30–70 ℃ are experimentally achieved, respectively.

Reflectance Transformation Imaging (RTI) is a computational photography technology that allows interactive manipulation of virtual lighting within two-dimensional images to enhance object surface information. RTI is based on the reflection characteristics of the surface of the object characterized by a specific parabolic function, of which the most classic is Polynomial Texture Mapping (PTM) based on biquadratic polynomial. Therefore, RTI is ideal for diffusely reflecting objects, and metal materials with specular reflection are nearly out of consideration before. In this article, several metal samples with different surface topography were selected to explore the potential of RTI for subtle traces on metal surfaces. The results show that, with the cooperation of different image enhancement algorithms, especially Specular Enhancement and Diffuse Gain, RTI can be well applied to the detection of metal, even metal with relatively strong specular reflection. And the No-Reference Structural Sharpness (NRSS) assessment is consistent with subjective perception, confirming that RTI not only helps to explore the 2D or 3D geometric texture of the surface of metal materials, but also plays an important role in understanding and spreading the cultural phenomena implied in the content on metal surfaces. Promisingly, RTI is expected to be extended to a wider range of fields.

The assembly gap between components is very vital for the evaluation of assembly quality of aircrafts. Due to the limits of gap size and operation space, the assembly gap needs to be indirectly calculated by the measurements of surface of components instead of plug gauge test. However, the surface constituted of point cloud is usually mixed with different types of noise ,which severely affects the evaluation of assembly gap. To remove these different types of noise simultaneously with high efficiency, a classified denoising method combining with an improved bilateral filtering and median filtering was proposed. Firstly, based on the principal component analysis, a new coordinate system was established to achieve the homogeneity of coordinates of point cloud. Then, an improved median filtering method on the basis of region segmentation (RSMF) was used to remove large-scale noise. Next, the fast bilateral filtering method based on threshold segmentation (TSBF) was proposed to remove small-scale noise. Finally, a measurement experiment of aircraft component was performed to verify the effectiveness of the proposed method. Experimental results showed that the proposed method could not only reduce measurement error including RMSE (Root Mean Square Error), but also improve SNR (Signal Noise Ratio) and PSNR (Peak Signal to Noise Ratio) of point cloud data.

A fiber loop ring-down (FLRD) magnetic field sensing system combined with frequency-shifted interferometry (FSI) was proposed and the effect of temperature on its performance was experimentally demonstrated. The FSI-FLRD technique measures the light intensity decay rate (called the ring-down distance) in the space domain instead of in the time domain. Compared with conventional time domain FLRD scheme, FSI-FLRD technique greatly reduces the cost due to only need of inexpensive continuous-wave laser and slow detection. The tapered single mode fiber surrounded by magnetic fluid (MF) was utilized to construct the sensor head for temperature measurement. Different temperature was generated due to the thermal effect of MF and the temperature was obtained by measuring the ring-down distance. The experimental results indicated that a sensitivity of 0.31 /(mm·°C) was achieved.

It is a key problem to get the high resolution and large field of view (FOV) of dynamic measurement in the display technique. In this paper, Charge Coupled Device (CCD) was synthesized with seams to be a display unit. A reconstruction method is proposed which combines the principles of synthetic aperture and phase retrieval to detect the 3D particle target with synthetic CCD. In the simulation experiment, the correlation coefficient of surface l is 0.73217 between the reconstructed image and the simulated particles target image. The pictures from the experiment depicts the particle is not only clearly visible at the normal region but also at the seams. The optical information can be recovered well of the high resolution and large field of view (FOV) with this method.

Previous studies on the use of photoacoustic spectroscopy (PAS) techniques in the measurement of CH₄ concentrations have been mainly limited to a single point measurement using only one PAS cell. A single point measurement will not satisfy the detection need, for instance in a natural gas transmission pipeline or an underground coal mine, gas explosion may occur at any point of the pipe or tunnel. Therefore, simultaneous multi-point measurements are required to obtain the gas distribution profile along the tunnel. Here we proposed and demonstrated a multi-point photoacoustic spectroscopy gas detection system based on the power compensation technique and fixed wavelength technique for CH4 detection. In the multi-point photoacoustic spectroscopy gas detection system, the gas absorption of the first PAS cell would make the measurement of the second PAS cell inaccurate. Therefore, we proposed a power compensation technique to compensate the absorbed laser power. In addition, a simple fixed wavelength technique was used in multi-point photoacoustic spectroscopy gas detection system to improve the signal to noise ratio (SNR). Finally, the experiment results showed that the two PAS cells achieved the minimum detection limit of 7.72 ppmv and 5.14 ppmv, respectively.

It is well known that achieving a robust visual tracking task is quite difficult, since it is easily interfered by scale variation, illumination variation, background clutter, occlusion and so on. Nevertheless, the performance of spatio-temporal context algorithm is remarkable, because the spatial context information of target is effectively employed in this algorithm. However, the capabilities of discriminate target and adjust to scale variation need to promote in complex scene. Furthermore, due to lack of an appropriate target model update strategy, its tracking capability also deteriorates. In the interest of tackling these problems, a multi-scale spatio-temporal context visual tracking algorithm based on target model adaptive update is proposed. Firstly, the histogram of oriented gradient features are adopted to describe the target and its surrounding regions to improve its discriminate ability. Secondly, a multi-scale estimation method is applied to predict the target scale variation. Then, the peak and the average peak to correlation energy of confidence map response are combined to evaluate the visual tracking status. When the status is stable, the current target is expressed in a low rank form and a CUR filter is learned. On the contrary, the CUR filter will be triggered to recapture the target. Finally, the experimental results demonstrate that the robustness of this algorithm is promoted obviously, and its overall performance is better than comparison algorithms.

Microchannel plate (MCP) is an important compact signal multiplier. The dynamic range of MCP is the key index influencing the detection quality of high flux, high energy and wide output linearity signals, but limited on the bulk resistance of MCP multiplier. Therefore, the problem of the bulk resistance-dependent of dynamic range transition in these devices needs to further continues to be investigated. In this paper, lead silicate glass microchannel plates in three different bismuth contents (0at%, 1at%, 2at%) were present. The bulk resistance decreased exponentially (from 6000MΩ into 15MΩ) with the increase of the content of bismuth in the lead silicate glass from 0at% to 2at%, meanwhile, the dynamic range was increased from 3.3×10² to 2.3×10³ . As the bulk resistance of MCP multipliers dropped from 150MΩ into 15 MΩ (dropped about 90%), while the dynamic range of that just extended by seven times, meanwhile, the dark current of that boosted almost ten times (from 1pA to 10pA). It implicated that the dynamic range of MCP multiplier was influenced not only by the charge replenishment abilities involved in the bulk resistance, but also by the joule heating effect-depended dark current variation concerned with the bulk resistance.

Infrared thermal camera is one of the important sensors in electron optic tracking system (EOTS) of UAV, and its parameters and performance directly affect the reconnaissance and surveillance performance particularly the combat efficiency in night of the UAV system. In this paper, we propose a method for determining the parameters of infrared thermal camera in EOTS of UAV. This method can easily determine parameters of infrared thermal camera from the aspects of working wavelength, sensors, lens size and focal length and so on. It realizes recognition distance of EOTS and can be used to guide the selection and integration of the infrared thermal camera in EOTS of UAV.

We propose a new optical engine that supresses stray light to address the problems of high stray background light and low star point simulation in traditional star simulators and offer an experimental proof that the new design provides higher contrast. Our design incorporates an optical collimator with a modulation frequency transfer function (MTF) greater than 0.6 at a cut-off frequency of 60 lp/mm that meets star point identification requirements. Experimental results show the background noise of the high contrast star simulator to be well suppressed, providing a 3.16-fold reduction in stray light compared with a traditional system.

Luminance gain is an important parameter to evaluate the light intensity enhancement ability of the low light level image intensifier assembly. The higher the luminance gain, the easier the receiver is to sense and recognize. However, luminance gain is not a directly measurable physical quantity. Thus, luminance gain measuring devices have non-standard specific properties. Based on the principle of luminance gain measuring specified in the standard, the structure and measurement methods of measurement devices are analyzed, the error and optimization methods of two major measurement methods are compared, and the distribution of the combined uncertainty of measuring luminance gain is studied. Then, an optimized measurement scheme of luminance gain of low light level image intensifier assembly is put forward. Based on this scheme, a comprehensive measurement uncertainty analysis is carried out and the calculated luminance gain measurement extended combined uncertainty is about 6.7% (k=2). The results are of great significance for improving the measurement accuracy of luminance gain of low light level image intensifier assembly.

In the process of fast automatic alignment of optical path in the integrated diagnosis system of host laser facility, there are six kinds of different optical target images captured. Because of the different characteristics of optical target, different methods for calculating the center of optical target image must be put forward, which increases the algorithm complexity of calculating the optical target center. In order to improve the efficiency and accuracy of fast automatic alignment of laser beam in the integrated diagnosis system, it is necessary and important to propose a method that can calculate all kinds of optical center of collimating images. In this paper, based on the synthesis of the characteristics of pinhole image, schlieren sphere image and common optical target image, a general method for calculating the optical target center of weak contrast collimating image is proposed. Firstly, multi-dimensional image cubes are constructed by using multi time-sharing images or neighborhood vectors, and the dimension of multi-dimensional data cubes is reduced by using NVPCA transformation to remove the correlation between the various dimensional images, separate and readjust the noise of original image; Secondly, the one-dimensional image data is classified by Kmeans, and the binary image is processed by the mathematical morphology operation to separate the laser target and background respectively; Thirdly, if the optical target image is a schlieren image, it need to obtain the background of schlieren image, the region of schlieren sphere, the edge of schlieren sphere respectively; If the optical target image is a pinhole image, it need to obtain the characteristic points on the circle contour of the pinhole image; Finally, the least square method is used to calculate the circle center of the pinhole image and the schlieren sphere image, and the center of gravity method is used to calculate the center of optical target. The experimental results show that this paper synthesizes the same and different characteristics of all kinds of collimating images, proposes a method to calculate the optical target center of weak contrast collimating images, improves the calculation accuracy of the optical target center of collimating images (less than 1 pixel) effectively, shortens the fast alignment time of the optical path of the integrated diagnosis system, and provides an effective guarantee for the measurement experiment of far-field focal spot of high power laser based on schlieren method.

It is known there are some problems such as the image suffers from excessive brightness or darkness under non-uniform illumination, loss of detail in the enhanced image, halo effect and so on. In this paper, an adaptive correction algorithm is proposed for these problems. Firstly, based on the illumination-reflection model, the illumination component was extracted by weighted guided filtering integrated with gradient information. Afterwards, the index of 2-d gamma curve is adjusted with the illumination component to realize the adaptive illumination correction. Finally, the output image pixels are produced with their adjacent pixels to enhance the local contrast of the image. The experimental results show three aspects. One is the improved weighted guided filtering can distinguish the edges adaptively and avoid the halo caused by excessive smoothness near the edges. Two is the clarity of enhanced image was improved by over 86.5% on average. At the same time, due to the reduction of color distortion, the ability to enhance image detail was improved by 18.5% on average.

As a typical atomically thin two-dimensional (2D) crystal, graphene-based photodetector has attracted tremendous attention in recent years due to excellent photoelectric performance of graphene. However, due to the unstable chemical properties of graphene, the guidance is lacking in the design of composites with other materials. The stability and repeatability of the device are also restricted by the material properties. Here, we have established a graphene simulation model and obtained the electrical characteristics of graphene which are consistent with the experimental results by using the method of equivalent design. We simulate the detection capability of graphene-silicon composite photodetector and graphene-germanium photodetector under different doping conditions for visible light (532nm) and near infrared wave (1550nm) respectively. The current-voltage characteristics of the heterojunction are obtained. Through this model, the proposed bottom gate adjustment mechanism makes the carrier transport of graphene/silicon-on-insulator (GSOI) photodetector adjustable. By adjusting the carrier concentration and distribution in the graphene channel and silicon through the bottom gate voltage, an enhanced built-in potential is obtained, which increases the device responsivity by 5-10 times to 10⁶ A/W. This model can be used to simulate the carrier distribution and current density when different gate voltages are applied, and the modulation law of this model is in good agreement with the experimental results. This work will be beneficial to the mechanism analysis and performance optimization of the graphene composite structure photodetector, which has important guiding significance for the experiment.

A robust multi-focus image fusion method is proposed to generate an all-in-focus image with all objects in focus by merging multiple images. The proposed method first estimates local focus maps using a novel measure of Gaussian model combined with joint bilateral filtering in HSV space. Then, a propagation process is conducted to obtain accurate focus maps based on a traditional natural image matting model that makes full use of the spatial information. The fused all-infocus image is finally generated by a focus-selected strategy. Experimental results demonstrate that the proposed method has state-of-the-art performance for multi-focus image fusion under various situations encountered in practice, even in cases with little edge information.

Aiming at the problems of target scale change, color similarity and occlusion during target tracking, this paper proposes a single target tracking algorithm based on the fusion feature of color feature (CN) and direction gradient histogram (HOG). Under the relevant filtering and tracking framework, the original RGB color space is mapped to the color attribute space to reduce the target color from being affected by environmental changes during the tracking process. The adaptive component dimensionality reduction through principal component analysis (PCA) method, features The number of channels drops from 10 to 2, and the cost of crossing different feature subspaces is increased by smoothing constraints. At the same time, the direction gradient histogram is extracted, and the feature map is calculated by kernel correlation filtering to obtain the correlation response map, and the maximum response value is found from the response map to determine the target position. 36 groups of color video sequences were selected on the OTB standard data set for experiments. The popular correlation filter tracking algorithm was compared. The experimental results show that the algorithm has high recognition accuracy and can be used in complex environments such as illumination changes, target occlusion and deformation. Stable tracking target.

By combing the technique of coherent detection with ghost imaging, we establish an experimental system for pulse-compression ghost imaging via coherent detection. The results have experimentally demonstrated for the first time the feasibility of ghost imaging via coherent detection. More importantly, it shows that even when echo power is only 5 pW, which is lower than three orders of magnitude compared with previous ghost imaging via photon intensity detection, an image with a spatial resolution of 0.8 mm can be obtained for the proposed method.

In order to better characterize the polarization characteristics of reflected light in oil spill monitoring by remote sensing on the sea surface and improve the accuracy of oil spill detection, the quaternion method is introduced into the oil spill detection system, and a kind of oil spill monitoring technology based on quaternion is proposed in this paper. Firstly, the quaternion matrix method for calculating the sea surface reflection polarization information is proposed, and the relationship between the characteristic parameters of quaternion and the refractive index of oil film is discussed. Then， four kinds of oils crude oil, soybean oil, lubricating oil and diesel oil are used as the experimental samples for oily pollution on water surface, and the characteristic quaternion parameters, s and p components of different samples are obtained through the reflection images at different polarization directions, so the identification of different oil products is realized.. The experimental results show that the quaternion polarization characteristics can be used as an important parameter for oil identification.

The false alarm generated by the background object in the complex sky background is a difficult problem in the detection of infrared small targets. The problem is particularly prominent when the target signal is weak. In order to effectively suppress the false alarm rate and improve the detection rate, a weak target detection algorithm based on infrared polarization map is proposed. Firstly, the data is collected by the infrared polarization detecting device, and then the polarization component map with the best image quality is analyzed. Finally, the filtering algorithm combined with Top- Hat morphological filtering and median filtering is used, and the global component threshold detection method is used to complete the polarization component map. Weak target detection. The experimental results show that the proposed method has higher detection rate and lower false alarm rate than the Top-Hat algorithm, maximum mean filtering algorithm, and LCM algorithm. An effective method for detecting weak targets in the background.

With the development of infrared surveillance technology, the short focus zoom LWIR lens has been paid to more and more attention. In this paper, dual fields zoom lens with the fields of 20°×25°/8°×10° are designed by two different methods. The main specifications of the lens include: aperture is 1 / 2, detector resolution is 640 × 512, pixel size is 15 μ m, temperature adaptability is -40 ° to + 60 °. And we compare the volume, weight, imaging transmission, transmissivity and economy of the lens by different methods. It is found that the re-imaging method has smaller volume, weight and economy. While another imaging method has better transmittance, so the imaging system can reach higher noise equivalent power at the same image quality

Different from the traditional mechanical gyroscope, fiber optic gyroscope (FOG) has the characteristics of fast start-up speed, high precision, small volume, and low cost, which has been widely used in the fields of sea, land, air and so on. With the wide application of FOG, the problem of its reliability has gradually emerged. Among all kinds of reliability problems, the problem of degradation of the input axis misalignment angle of high-precision FOG with the time is particularly prominent. If it cannot be solved as soon as possible, this problem will seriously affect the actual use of highprecision FOG In this paper, a fault tree is established, which takes the degradation of the input axis misalignment angle of high-precision FOG with time as the top event. The physical and chemical factors and manufactucoil process parameters that cause the degradation of the misalignment angle are analyzed. Through the study of the degradation mechanism, it is concluded that the change of the size stability of the fiber coil is one of the main factors that cause the degradation of the misalignment angle with time. Then we design and build a special testing system for the misalignment angle of highprecision FOG. The degradation mechanism of the misalignment angle of high-precision FOG is verified by experiments. The theoretical analysis and experimental results show that the stability of the size of fiber-optic coil changes with time, which will have a very bad impact on the reliability of the misalignment angle of FOG. Finally, through the design and process optimization, the reliability of the misalignment angle of gyro can be improved.

As a new generation of INS products, high-precision FOG has been widely used. High precision fiber coil is the sensitive core of high precision fiber optic gyro. Its performance and reliability index determine the performance and reliability of high precision fiber optic gyro. In this paper, the key process methods and parameters that affect the performance and reliability of optical fiber coil are studied. Firstly, the process flow chart of high-precision fiber optic gyroscope is drawn, and every process link of high-precision fiber optic coil manufacture is deepened. Through theoretical analysis, the key process and process parameters affecting the performance and reliability of high-precision fiber optic coil are identified. Then carry out the corresponding process validation test at the same time, through the analysis of the test results to verify the previous theoretical analysis. Then the optimization design and reliability growth research are carried out for the key process, and the control of the key process is strengthened to improve the performance and reliability index of the coil from the process link.

Combining support vector machine with infrared spectrum analysis, the infrared spectra analysis model of mixed gas is established. Taken the mixed gas containing hydrocarbon as an example, the selection and optimization of model parameters is researched through experiment. Beginning with the SVM analysis model, infrared spectrometer and spectra data, spectra analysis band and spectrometer scanning interval, the detailed study of how the parameters like the types of kernel function for the SVM analysis model, penalty factor C, method of spectra data preprocessing, spectrometer scanning interval, spectra analysis band, etc. Affect the analysis result are carried out. The experimental results show that under the condition of the SVM analysis model is determined, a reasonable selection and optimization of the SVM analysis model parameter can improve the accuracy of the analysis results, and it has a practical application value.

Lead sulfide colloidal quantum dots, similar to the nanoscale crystals of most semiconductor crystals, are available in a variety of sizes, shapes, and compositions as well as to make different chemical molecular ligands to modify the surface of the quantum dots and to fabricate functional optoelectronic devices on a variety of substrate materials. The combination of silicon and colloidal quantum dots enables the fabrication of silicon-based compatible quantum dot optoelectronic devices over a wide range of applications. In this paper, the effects of channel doping concentration and channel length on the performance of silicon-based CQD/Si photodetectors are calculated and analyzed from the simulation method. The results show that a suitable doping concentration and a short channel length can improve the performance of the device, which provides a simulation basis for the fabrication of silicon-based compatible arrayed colloidal quantum dot photodetectors.

The infrared characteristics of the target is very important for target recognition. The infrared radiation characteristics are different between damaged solar panel and normal working solar panel. It is the characteristic position for the solar panels’ nondestructive fault testing. In this paper, the temperature model of solar panels is established, and the numerical calculation is carried out. The calculated results and the measured data matching well verify the correctness of the calculation model. The results also show that the temperature of solar panels which are damaged are higher than the temperature of those who are normal working. On the basis of the temperature calculation, the infrared radiation characteristics of solar panels are calculated .The result shows also that the infrared radiation intensity of those damaged solar panels are higher than the normal working solar panels. This provides the theoretical support for the infrared nondestructive testing of solar panels.

Based on the principle of compressed sensing, a single-pixel imaging system (single-pixel camera) was built in our laboratory, and our single-pixel camera can successfully acquire the picture of imaging scene of a distant building. However, the process of single-pixel imaging is easily disturbed by noise. In order to reduce the interference of noise in the imaging results as much as possible, we apply the method of adaptive dictionary learning to denoise the reconstructed image in our single-pixel camera. The block-coordinate decent idea and K-SVD algorithm are used to complete the dictionary learning by training on the noisy reconstructed image in the method of block-by-block scanning, which doesn’t require the extra corpus of image patches, so the method of dictionary learning is self-adaptive to the imaging picture that need to be denoised. The overcomplete DWT (discrete wavelet transform) dictionary is applied as the known initial dictionary that will be trained in the adaptive dictionary learning. The effectiveness of the adaptive dictionary learning has been demonstrated by the simulation experiment and the practical imaging experiment, the method of adaptive dictionary learning based on the initialization of overcomplete DWT dictionary effectively reduces the imaging noise and improves the image reconstruction accuracy and imaging quality.

The traditional operating distance model of infrared imaging system is only valid under the condition that the background is single and invariable. For the sea surface targets, radiation characteristics are affected by the sea-sky-line, specular reflection, sea clutter and so on, resulting in a large calculation error of operating distance if traditional operating distance model was used. In this paper, the operating path of infrared imaging system is divided into chaotic path and methodical path. The chaotic path is the operating path with a certain slant distance from the sea surface target, the infrared radiation propagation characteristics of the target in this path are uncertain, multivalued and volatile, showing certain chaotic characteristics. The methodical path is from observation point of infrared imaging system to the terminal point of chaotic path, the infrared radiation propagation characteristics of the target in this path are certain regularity and linearity, which can be estimated by using the traditional operating distance model. The main work of this paper is to subdivide and characterize the infrared radiation characteristics of the sea surface target in the chaotic path, and finally linearize it in a specific situation, convert the radiation energy of the sea surface target to the starting point of the methodical path, and at last calculate the operating distance according to the traditional operating distance model of infrared imaging system. This method is innovative in theory and can give a reference to the research of infrared system detection range under complex backgrounds.

The emergence of HTV-2, X-51A and other near space hypersonic vehicles poses a new challenge to radar target tracking. Near space hypersonic vehicle has the characteristics of high speed and high maneuverability, but the tracking performance of conventional tracking algorithm is low, which can not meet the operational requirements. In this paper, the technical difficulties of near space target tracking are described. By analyzing the motion characteristics of near space hypersonic vehicle, the dynamics model of HTV-2 is established, and a tracking method based on interacting multiple model (IMM) for near space vehicle is proposed. At the same time, considering the unscented Kalman filter (UKF) algorithm has a better filtering effect on strong nonlinear object. The simulation results show that, the interact multiple model－unscented Kalman filter (IMM-UKF) algorithm is better than the single model EKF algorithm.

In the process of crack identification for round logs, conventional edge extraction cannot effectively suppress noise because of the tree's annual ring lines and the similarity between the burr noises during cutting and the gray level of the target. Therefore, it is no easy to extract the target crack. The method of continuous gray-scale transformation enhancement is put forward in this thesis to increase the difference between the gray level of the background pixel and the gray level of the target so that can obtain an ideal pre-processed image. In the process of image preprocessing, the method of continuous gray-scale transformation enhancement is applied, that is to combine the gray-scale transformation enhancement and the non-linear filtering process so that can realize the preprocessing of the original image. The gray level difference between the extraction target and the background is increasing under the premise of preserving the image-extraction features. In the extraction process, the extracted target crack image is obtained through utilizing the localization minimum in mathematical morphology and then the compound morphological algorithm is designed based on the basic algorithm of mathematic morphology so as to obtain the target crack image which is connected by the edge curves. Results The MATLAB image processing algorithm is used to simulate each step of the method. The results show that the extracted target crack images are ideal. The mentioned algorit can not only ensure the integrity of the extraction target, but also can suppress the noise very well so that can satisfy the needs during the extraction of complex background images, especially the images with little difference between the background gray level and the extraction target gray level.

Optical antenna is an afocal telescope system which used to transmit or receive optical signals in free-space laser communication systems. In designing a high performance optical antenna, the challenging tasks including achieving of very compact and light weight optical configuration, appropriate exit pupil distance, high transmission efficiency, and wide temperature working range. The formulae of Gaussian parameters between the objective and eyepiece are derived by using the ideal imaging process. A first-order optics analysis of the optical antenna is given, which allows optics engineers to pre-define a suitable optical layout before performing design optimization and analysis. Design examples of off-axis reflective optical system are presented.

With high time resolution and wide coverage, GF-4 camera can achieve high-frequency repeated monitoring over a wide range of areas within a certain period of time. However, there are some problems of its sequence images, such as shaking, non-prominent ship targets and insufficient brightness in sea area, which are not conducive to the detection of moving ships. Thus, a hierarchical registration algorithm combining SURF and SIFT, and an image enhancement algorithm combining wavelet threshold denoising and Top-hat filtering was proposed to solve these problems. Matlab’s experiment proves that the processed image can almost only contain the highlighted target and the black background, which improves the accuracy of subsequent ship monitoring.

Aiming at the problem of poor detection of slow speed targets in sea surface under complex sea conditions for airborne maritime surveillance radar, this paper first analyzes the distribution characteristics of sea clutter in different sea conditions, studies focus on inter-scanning coherent processing, non-coherent processing and frequency agility processing. Finally, the results of three methods, such as non-coherent accumulation, constant frequency coherent accumulation and fixed frequency non-coherent accumulation, are analyzed and compared under different sea conditions. It is concluded that under low sea conditions, the method of coherent accumulation and increasing the number of pulse accumulation can effectively detect the slow target of the sea surface.

In order to solve the difficult problem of unmanned air vehicle(UAV) target detection in visible light images under complex sky background, this paper proposes a UAV target detection method based on frequency domain transform. First, the B-channel in the image LAB space is used to extract the sky and cloud boundary images, and then the image feature channel is used to construct a quaternion function. Secondly, Fourier transform is performed on the quaternion function to extract the amplitude spectrum and phase spectrum, then the amplitude spectrum image is subjected to multi-scale decomposition using wavelet transform in the frequency domain, the amplitude spectrum image of each scale and the phase spectrum image are combined by inverse Fourier transform, and the evaluation function is used to obtain the best scale image. Finally, the best-scale image and the boundary image are normalized to make a difference to obtain the final detection result. Experimental results show that the algorithm can effectively detect UAV targets under complex cloud background.

The most important part of visible-infrared image characteristic simulation, is the degradation simulation of the image in the actual turbid atmosphere. Aerosols, molecules, turbulence, and imaging devices in the atmosphere are the dominant factors in the image degradation process. These effects occur simultaneously in the real-world atmosphere. According to the equivalence principle, the MTF of molecule can be calculated, whose results show that the absorption and scattering effect of molecule cannot be treated easily as a dampen on image contrast but is as important as that of aerosol. At and near some absorption bands, molecule MTF at high spatial frequency is smaller than aerosol MTF at high spatial frequency; at weak-absorption and strong scattering bands, molecule MTF is comparable to, even bigger than, aerosol MTF at high spatial frequency. In turbulence free atmosphere, the dominant factor that affects the image quality is determined on the dependence of weather and wavelength. At molecular absorption lines, the effect of molecule is dominant, whereas, at visible bands, the effect of aerosol is comparable to, even more important than, that of molecule. When the turbulence is considered, the result is a little complex. If turbulence blurs the image seriously and the cut-off spatial frequency is much smaller than that of overall turbid atmosphere MTF, the image degradation is dominant by the turbulence. In addition, the cut-off spatial frequency of aerosol MTF or molecule MTF is generally smaller than 2 cycles/mrad even 1 cycles/mrad, which means that aerosol MTF or molecule MTF will be dominant factors for imaging degradation when the turbulence is weak enough in atmosphere.

In this work, a new type of optical fiber pressure sensor based on Frequency Modulated Continuous Wave (FMCW) laser interference technology was designed and produced. The sensor consists of a diaphragm-type Fabry-Perot (F-P) cavity structure, which is made of 316L stainless steel with excellent elastic properties. The deformation of the diaphragm occurs under the gas pressure, which causes the length of the F-P cavity to change. The FMCW laser interference technique was employed to demodulate the change of cavity length. The experimental results show that the linearity of pressure and cavity length can reach 0.99993 whin the range of 0~600 kPa, and also verify that the pressure sensor has good repeatability and stability.

A mid-infrared TDLAS sensor near 2.5μm was designed for time-resolved measurements of temperature and water vapor partial pressure at the nozzle exit of a laboratory-scale hybrid rocket motor. Several previously used H₂O transitions within 2.4-2.9μm were thoroughly investigated, and a line-pair containing three transitions (4029.52 cm^-1 , 4030.51 cm^-1 and 4030.73 cm^-1 ) was selected for the optimal overall properties like strong absorbance, sufficient temperature sensitivity, single laser scan, high immunity from the ambient H₂O transitions and low measurement uncertainty affected by temperature over the range of 1500K-2500K. Firing tests were conducted on an oxygen/paraffin-fueled hybrid rocket motor operating at oxygen/fuel ratios (O/Fs) of 3.10, 2.77 and 2.88, corresponding to average combustion pressures of 1.91MPa, 2.09MPa and 2.38MPa. A distributed feedback (DFB) laser tuned repetitively at 2kHz was used as the light source, and simultaneously the transmitted spectra were detected at a 2MHz sampling rate. Finally, a 4.5ms time-scale variations of temperature and H₂O partial pressure were captured by TDLAS sensor. Uncertainty analysis was made in detail based on average temperature (1929.8K, 1926.5K, and 1990.7K) and average H₂O partial pressure (0.237MPa, 0.253MPa, and 0.285MPa), leading to temperature uncertainty of around 2.24% and partial pressure uncertainties of around 3.80%, 3.79% and 4.04% respectively. The time-resolved measurement results and small measurement uncertaintiesindicate that TDLAS has the potential to evaluate the combustion performance of hybrid rocket motor

In this paper, a novel optical method is proposed to effectively double the sensitivity of FOGs. Two fiber polarization combiner/splitters (PCS) are added in the traditional FOG optical path, which are able to either combine two orthogonal polarizations transmitting at two different PM fibers into the two orthogonal axes of one PM fiber respectively or split two polarizations transmitting at the two orthogonal axes of one PM fiber into two polarizations to transmit at two different PM fibers respectively. Through the specific placement and coupling of these two fiber polarization combiner/splitters, the incident light can transmit twice along the light path of FOG. The experimental setup is established and the method is verified experimentally. The results show that the proposed method will effectively improve the sensitivity and the signal-to-noise ratio (SNR) of FOG without increasing the length of fiber coil, which is very benefit to improve the technology of FOGs, as well as the miniaturization of FOGs.

Three-dimensional imaging lidar is a new type of active detection technology, which can obtain target spatial information accurately and quickly. It has a wide application in the fields of target detection and recognition, scientific research detection, mapping and navigation, etc. Three-dimensional imaging lidar has many modulation modes, among which the non-scanning three-dimensional imaging lidar based on polarization modulation has the advantages of long measurement range, high measurement accuracy, fast imaging speed and no motion artifacts, which is one of the hot research directions in this field. In view of this technology, this paper analyzes the principle of polarization modulation imaging by Jones matrix calculation, and obtains the geometric relationship between the imaging illumination values in four polarization directions and the polarization modulation phase delay of Pockels Box. Then, a set of laser radar optical system based on polarization modulation imaging is designed by using optical software. The analysis results show that the designed optical system has good imaging quality, clear target edge imaging and can distinguish independent square targets with a diameter of 1m; The light outside the field of view of the strip light source does not enter the detector, so the system has less stray light and less imaging distortion; At last, the input phase delay δ of the Pockels box in the simulation model is randomly set to different values, and the illumination values of four polarization directions are obtained by imaging simulation, so that the phase delay δ′ of the Pockels box is inverted. The results show that | δ′-δ | ≤ 5.2× 10^-6 λ, which proves that the polarization modulation method is correct.

Laser active imaging uses a pulsed laser to illuminate the target. Through imaging detection of the target's reflected light, more detailed information of the target can be obtained. It can be used in severe weather conditions, at night and in the light of fire or smoke to monitor the target and attitude measurement. This article introduces the working principle and research status of four active laser imaging methods, including commonly used point scanning imaging, continuous light imaging, fringe tube imaging and APD imaging.

Among pathological images, there are usually some typical patient structures. For these typical images, template matching can be used for diagnosis, thereby the workload of doctors can be reduced. In this paper, pathological images of breast masses are studied in depth. A typical breast masses pathological template is established. Three template matching methods, correlation coefficient matching, correlation matching method as well as square difference matching method are used for experiments separately, and the timeliness and effectiveness of them are evaluated in terms of computing time and matching accuracy. The experimental results show that among these three template matching methods, the accuracy of correlation coefficient matching is the highest, and it is able to overcome the interference of angel rotation and noise. However, it is also consumes the longest time.

Using the output of the photodiode as a current source, and then introduce an operational amplifier to convert the photo-generated current to voltage becomes a typical photodiode amplifier circuit. The seemingly simple circuit faces a difficult problem when detecting weak photo-generated current: the photosensitive surface of the diode and the resistor using for current-voltage conversion need to be enlarged as much as possible to increase the output amplitude. However, this will result in a reduction in the bandwidth of the photodiode amplifier circuit, a reduction in stability, and an increase in noise. It was therefore decided to introduce software to assist in circuit design for the improving of performances of circuit. To assessing the applicability, the photodiode amplifier circuit was matched with other optoelectronic devices such as laser driver, temperature controller and lock-in amplifier, and employed in wavelength modulation spectroscopy (WMS) for NH3 detection at its P(6) absorption line belonging to the ν3 + ν4 combination band. Under the experimental conditions of room temperature (296 K), normal pressure (1.01e5 Pa) and an effective absorption path length of 10 cm, the experiments were conducted using five samples with their concentrations ranging from 500ppm to 1%, and a 1.3ppm detection limit was deduced from the linear regression analysis. The results not only indicate that the photodiode amplifier circuit has great potential in other trace material detection applications based on absorption spectroscopy, but also reflect from the side that the method of using simulation software to assist circuit design is highly feasible.

In this paper, based on the collaborative integration and modeling calculation of single-point Laser Doppler Vibrometer (LDV), a 3D oblique non-contact dynamic attitude measurement system is proposed. By using two groups of three-dimensional measurement devices which consist of three LDVs, our system can measure the translation and rotation of the measured object. The experiments are carried out under line vibration environment, and the elicited results are compared to those elicited using fiber optic gyroscope (FOG), which verified the feasibility of our system.

Schlieren method is an effective method to measure the high dynamic range far-field focal spot of high power laser at present, it has been used in the comprehensive diagnosis system of the host facility of the national large scientific device. However, the background subtraction algorithm is used for denosing process of the main lobe and side lobe images in the experiment, which could not effectively remove the noise under the condition of strong radiation. In order to improve the reconstruction accuracy and reliability of far-field focal spot measurement based on schlieren method, there are two aspects have been proposed and optimized in this paper as follows, the mathematical model for the measurement of high dynamic range far-field focal spot is constructed, the noise of images captured by detected CCD is removed effectively. Firstly, the mathematical model of laser focal spot measurement is constructed, which provided a theoretical basis for the measurement of far-field focal spot based on schlieren method. Secondly, the denoising algorithm based on Convolutional Neural Network (DnCNN) is introduced into the denoising of main lobe and side lobe CCD image, which can effectively remove the noise of different levels (0-75db) of main lobe and side lobe CCD image.The experimental results show that the noise signal of main lobe and side lobe images are removed by using DnCNN algorithm, and the best reconstructed image of far-field focal spot of high power laser is obtained when the influencers of amplified noise of main lobe image is reduced greatly. After the above optimization and improvement, it will meet the requirements for accuracy and efficiency of the measurement of high dynamic range far-field focal spot.

PS plate (presensitized plate) is a new type of plate used in the fast printing industry. Due to the "squeaking" caused the production process, it will seriously affect the printing quality. This kind of helium detection device is currently entirely monopolized by large foreign companies, the price is expensive, and it is also not suitable for the conditions of domestic enterprises. The development of this device will fill the gap in the domestic field. At present, most domestic companies adopt manual inspection methods for naked eye observation. Because of the subjective nature of human beings, it is impossible to quantitatively analyze. The PS version quality report provided by the device development provides data basis for later-stage sheet and stacking work which can realize Intelligent production.

In recent years, the number of small unmanned aerial vehicles (UAVs) present in the airspace around us has increased dramatically. Unmanned aerial systems were traditionally associated with military use, however, the increase in drone activity is far from exclusively being generated by increased military use. The increase in micro-UAVs has made it a tough challenge for the detection of UAVs. The demand for measurement and analyze of radar characteristics of UAVs is increasing significantly. This paper focuses on the Radar Micro-Doppler signatures of small UAVs, and experiments were conducted to support the analysis. In this paper, the characteristic of rotor blades were shown, and the doppler signatures of a real micro-drone were measured by a pulse radar system, and the short time Fourier transform was used to analyze the micro-doppler signatures from the data collected.

Based on basic principle of infrared radiation, aiming at infrared radiation characteristics of aircraft in different altitude, speed and attitude, a dynamic measurement method of infrared radiation characteristics based on flight parameters is proposed, and the infrared radiation intensity of a typical aircraft is simulated and analyzed. A typical aircraft is simulated and analyzed. These conclusions provide a basis for stealth aircraft design and the IR guidance air-defense weapon system design and application.

Orbit prediction is the premise of consistency transfer calibration of space radiation standard. By using the satellite's orbital elements and the calibration sites information, the time and location of the two satellites are predicted with the joint observation constraints. The orbit prediction process of the consistency transfer calibration of space radiation standard is briefly introduced, and the satellite space position calculation method for orbit prediction is studied. By selecting TERRA and Landsat8 as the radiation reference satellites, GF-1, GF-2, GJ-1, etc. as the satellites to be calibrated and analyzing the constraints such as the observation time interval and the observation angle difference that limit the joint observation cross frequency, the analysis of the cross-matching constraint mechanism is done with the six calibration sites. The simulation and analysis results are helpful for the optimization design of the standard satellite orbit with the cross frequency of twice a month, and could provide technical support for the high-frequency and high-precision in-orbit radiometric calibration for Chinese satellites.

Super-wide FOV lens has a short focal length, making the detection distance shorter. To suit the need of long detection distance, the inversed telephoto structure is used. In this paper, we designed a super-wide FOV camera system with a negative-positive inversed telephoto structure, whose FOV is 95°×71.25°.And optical aberration were analyzed detailedly, the structure of foreside and backside were made certain respectively, based on these, optical optimum design was accomplished.The result shows that in the entire field of view, the MTF at 60lp/mm is more than 0.4, the diameter RMS of spot diagram dispersion circle is less than 5 microns and the maximum distortion is less than 5%. The result shows that it meets the requirement of system well.

In recent years, requirements such as over-the-horizon, high resolution, large field of view, wide spectrum and dexterity have pointed out new directions for the development of optical imaging systems. Common optical surfaces such as planes and spherical surfaces have been difficult to meet the requirements of optical indicators. It is often necessary to use multiple high-precision complex optical surfaces such as high-order aspherical surfaces, off-axis aspherical surfaces, non-rotational symmetric curved surfaces, microstructured optical array surfaces and freeform surfaces. This brings processing, testing and adjustment in terms of problems, the monolithic multisurface optics as a new type of optical element are increasingly attracting researchers' attention.

Images captured from outdoor scenes are usually suffer from the atmospheric scattering and absorption, these natural phenomenon of hazy cause low contrast, poor quality of images, which severely affect their subsequent applications. Despite numerous image dehazing methods have been developed to improve the quality of hazy images, effective image dehazing method remains a challenging problem. In this paper, we propose a new method for retina color perception-based image dehazing, which considers the light response model based on the retina color perception mechanism, and the transmission estimation model combines with the dark channel prior. Specifically, an adaptive light response model is designed based on stimulus perception of retinal ganglion cells and lateral geniculate nucleus (LGN) neurones in the receptive field; and a new transmission estimation model combine with the dark channel prior is developed, thus we propose a retina color perception-based image dehazing method. Moreover, to validate the effectiveness of the proposed dehazing method, we test the method on various synthetic hazy images, and compare with several state-of-the-art image dehazing methods quantitatively. Extensive experimental results shown that the proposed method could improve the clarity of the hazy corrupted images, and the details of dehazed images could be protected moderately.

As a sort of low-altitude remote sensing technology, UAV remote sensing possesses the advantages that satellite remoting sensing and traditional aerial remote sensing do not have. It can be effectively used as an indispensable supplementary means for obtaining remote sensing images. The paper develops a set of image processing systems basing on the characteristics of UAV remote sensing platform, which can provide digital images with high quality as well as high resolution. In this paper, firstly, a series of preprocessing such as noise reduction and enhancement will be performed on the obtained data. Then, the feature points in the images will be extracted, the potential image matching pairs will be set up and the mosaicking process will be completed. Finally, the UAV remote sensing images will be segmented and the road extraction of UAV remote sensing can be realized. Experiments show that with known camera calibration parameters, the accuracy of image mosaic is high. Also, roads are extracted from the sequence of images taken by the UAV continuously, with an accuracy of more than 90%.

With the progress of science and the development of society, the study of material composition becomes more and more important, and the identification of material composition is mainly to distinguish the spectral information of different substances. Spectrometer is an important optical instrument, which combines the optical method with modern electronic data processing technology, and accurately analyzes the structure, composition and content of the target substance by obtaining its spectral information. At present, it has been widely used in some important scientific fields such as field exploration, material detection and space-borne analysis. In this paper, a new type of spectrometer based on periodic array structure is proposed. The spectrometer modulates the phase of the incident light by using the small hole array structure with the diameter of sub-wavelength. By using the diffraction effect of light, the light intensity distribution of the incident light is directly recorded by Charge Coupled Device（CCD）according to different wavelengths. After passing through the diffraction aperture array, the light intensity distribution is recorded again, and the transmittance coefficients of the diffraction aperture array for different wavelengths are obtained respectively. Finally, the transmittance matrix of diffraction aperture array for incident light is obtained, and the spectral curves of different incident light can be obtained by data processing algorithm according to the transmittance matrix. The main optical device of the spectrometer is diffraction aperture array, which is a metal film coated on a transparent substrate made of resin material. A series of aperture arrays with diameters of 2-78 microns and 10×10 are processed on the metal film by micro-nano processing technology. Any aperture has different diameters, and the apertures are periodically arranged on the metal film. By using diffraction effect, the incident light with different wavelengths will produce different light intensity distribution on CCD. Combined with data processing method, the incident light can be obtained. Compared with the traditional spectrometer, the new spectrometer has no moving parts in the system, which improves its stability, and has the characteristics of fast data processing, small size, low cost and high spectral resolution. In this paper, the theoretical analysis, simulation and experimental verification of the new spectrometer are carried out. The results show that the new spectrometer has obvious advantages compared with the traditional spectrometer and has broad application prospects.

In frequency-modulated continuous-wave (FMCW) ladar, researchers usually linearize the output of the ladar’s tunable laser by a reference interferometer for high-accuracy measurement, but the reference interferometer need to be precisely calibrated. The H¹³C¹⁴N cell is a universal tool to complete the calibration. Here, we use the optical frequency comb instead of the traditional HCN gas cell to calibrate the path-length difference of the reference interferometer for higher precision. In this article, we present a detailed introduction of the experiment for path-length difference measurement and data processing method. The finally experimental results show that this method can give a micron precision, the standard deviation is 2.560e-5.

It is well-known that the synchrotron radiation light source (SRLS) possesses the unparalleled superior characteristic compared with other light sources. Therefore, the investigation and development of the advanced SRLS has always been a major international hotspot. At the same time, there are more and more researcher are actively devoted to construct the new generation of light source in China. For example, the Hefei Advanced Light Facility (HALF) in University of Science and Technology of China (USTC), etc. For HALF, in order to provide machine researchers and users with an intuitive reflection of the quality of the light source, it is necessary to develop a synchrotron light spot measurement system which is employed for real-time and non-destructive monitoring the synchrotron light spot size. According to the pre-designed parameters of HALF, which corresponding to the characteristic wavelength is 0.5 nm. Obviously, the general optical material cannot be satisfied the imaging requirements in this wavelength. Furthermore, the transverse profile of the synchrotron light spot of HALF is need to be less than 10 μm, of which puts forward higher resolution requirement for the synchrotron light imaging measurement system. In view of the above-mentioned characteristics, in this paper, we concentrate on improving a monitor based on Fresnel Zone Plate (FZP) for measuring the synchrotron light spot size. In terms of the numerical simulation and theoretical analysis of the diffraction of the light source, it is shown that the simulated results are in good agreement with the pre-designed values.

In order to study the temperature distribution characteristics of near-space hypersonic vehicles, the numerical calculation method was adopted to simulate and analyze the aerodynamic heating effect of hypersonic vehicles in this paper. The simulation process consisted of model establishment, mesh division, and CFD numerical calculation. Based on the simulation results, the temperature distribution characteristics of hypersonic vehicles and the influence of flight speed on the body temperature were analyzed. The results show that hypersonic flight will increase the body temperature significantly. The temperature at the stagnation point of the warhead can reach up to 2185K, and the temperature in other areas along the axis of the projectile gradually decreases. As the flight speed increases, the temperature of the vehicle body rises rapidly. When the flight speed is at 6.58Ma, 8Ma, 9Ma, and 10Ma, the temperature at the stagnation point of the missile body can reach 2185K, 3172K, 3792K and 4772K respectively

Image information for single polarization parameters is weak, low contrast and the common visible light intensity image detail fuzzy problems, in order to further improve the target detection of polarization imaging detection system identification capability, put forward a kind of based on the sampling of shear wave transformation under visible light image fusion algorithms, intensity and polarization parameters can effectively improve the identification of targets in complex background. Polarization degree of the image and visible light intensity image using the sampling shear wave transformation under the decomposed high frequency subband and low-frequency subband, then low frequency subband image fusion rules design based on region distance energy weighted algorithm, and the high frequency subband image fusion rule is designed to combine guide take large filtering area of energy, will eventually high low frequency subband image by the NSST finally fused image is obtained by inverse transformation refactoring. By comparing the fusion results of this algorithm with those of other methods that adopt the same decomposition transformation method but choose different fusion rules, the experiment proves that this algorithm not only has the best visual effect, but also has the advantage in the objective evaluation index value.

Anomaly detection with visual information by distributed deep learning is proposed in the paper. First, visual anomalies are defined in a special application domain, which are very important and critical for safe operation. Secondly, deep convolutional neural network is chosen as detector for visual anomalies. Thirdly, detection results from different visual sources are fused to get higher accuracies and lower false alarm rate. Experimental results demonstrate that the visual anomaly detection framework proposed can achieve high performance and provide satisfactory security assurance.

In this paper, a TV component of 16 times visible light continuous zoom lens is presented based on 1280×1024@2.7 micron COMS device. The system adopts 4 groups of 12 lenses to achieve continuous zoom, the former fixed group consists of 1 set of adhesive lenses and 1 single lens, the zoom group use 2 single lenses and 1 set of adhesive lenses, the compensating group use 3 single lenses and 1 set of adhesive lenses, and the latter fixed group consists of 2 sets of adhesive lenses and 1 single lens. After selecting three focal length positions for calculation and simulation, when the focal length are 5mm, 40mm and 80mm, the MTF of each field of view is above 0.3 at 120lp/mm, the total length of the system is less than 200mm, the field Angle is 1.6°×24.1°, F# is 4, and the focal length can be up to 5mm-80mm. The root-mean-square value of the dispersion circle size of each focal length position of the 16-fold visible light continuous zoom TV lens is less than the size of one pixel, so the image quality of each focal length position of the system is better. When the system is in the short focus position, the variable doubling group and the compensation group are at two ends, and the distance between them is the largest. When the system changes from short focus to long focus, the variablesize group and the compensation group approach to the middle, and the distance between them becomes smaller. In the process of system zoom, the image quality is clear, the CAM curve is smooth and suitable for processing. Continuous zoom TV lens all adopt spherical mirror with the characteristic of manufacturing easily, and the glass materials of the system are all taken from Chengdu bright glass library that the materials are easy to obtain. The zoom curve of the continuous zoom lens is smooth and the image is clarity. The distortion of the full view field is less than 3%, which meets the system requirements of the TV lens components.

A homemade temperature controller, which features with robust and inexpensive, is designed for actively adjust and stabilize the operating temperature of the tunable semiconductor diode laser. The accuracy of this temperature controller is +/- 0.05K, which corresponds to 0.035 cm-1 in wavenumber for our selected laser diode with a center wavenumber of 6045.949 cm-1. This temperature controller uses the STM32H743 (an ARM processor) as the control core, and introduces a capacitive touch screen to realize the visualization of experimental data and the intuitive of human-computer interaction. Through the simulation using software, the performance of conventional PID algorithm and adaptive fuzzy PID algorithm are compared and analyzed. Based on this, the fuzzy PID control strategy is determined and implemented through programming. In order to improve the efficiency and the stability of the ARM processor, the temperature control process is divided into several threads such as temperature acquisition, control quantity calculation, PWM output, temperature data/curve display, and SD card storage. The RTX5 operating system is ported to manage the cooperative operation of the above tasks and the communications between them. Using wavelength modulation spectroscopy, measurements of methane concentration are carried out in a laboratory controlled environment (P ∼ 1 atm, T ∼ 296 K) to validate the proposed temperature controller. In a stainless cell, the experiments are carried out using five samples with their concentrations ranging from 1000ppm to 2%, and a 5ppm detection limit is determined by nonlinear regression analysis, with a regression coefficient as high as 0.9979.

Microchannel plate (MCP) is two dimensional arrays of microscopic channel electron multipliers. In this study, oxide thin films such as zinc oxide doped with aluminum oxide (AZO) as conductive layer and aluminum oxide (Al2O3) as secondary electron emission (SEE) layer were prepared in the pores of MCP via thermal atomic layer deposition (ALD). The bulk resistance of MCP in the suitable range (about 40~100 MΩ) was obtained by adjusting the percentage of zinc oxide (ZnO) cycles and the nano-oxide thin film thickness. As the tested voltage increased a behavior of negative temperature coefficient of the film was observed. After 5 days of continuous loading with 800 V bias, the resistance nearly doubled and stabilized. For the conduction mechanisms, the I-V curve obeys the ohmic law at low voltage region and the trap-controlled space-charge-limited conduction mechanisms as the applied voltage continue to increase. High gain performance (24000 @ 800 V) can be obtained by depositing only aluminum oxide film on traditional reduced lead glass microchannel plate. Meanwhile after the deposition of aluminum oxide film, the dark current density of the microchannel plate would increase to 1.8 picoampere in 15 minutes at DC 1000 V. Because of blocking effect the dark current density will drop to 0.03 picoampere in 50 minutes by depositing zinc oxide film with thickness of 4 nm before the SEE layer, and the resistance of MCP will reduce when the thickness of zinc oxide exceeds 4 nm.

For traditional camera calibration methods, the calibration accuracy of camera parameters is highly dependent on the feature extraction of the control points of the calibration target. However, due to problems such as perspective and lens distortion, the commonly used checkerboards, circular dots, and other calibration patterns will inevitably undergo large deformation, resulting in a serious impact on the accuracy of the extraction of control points. To solve this problem, a camera calibration method based on active phase calibration targets is proposed. The method uses the MI Pad that can automatically display the structured light patterns as the calibration target. Compared with the method of using a checkerboard calibration target, this method does not require pattern detection and avoids the process of manual marking. Also, based on the phase information of the structured light gratings, a large number of dense, high-precision control point coordinates can be obtained even in areas where the image edge distortion is severe. Not only is it highly automated, but it is also suitable for defocused cameras. Experimental results show that the reprojection error of our method is only onetenth of the comparison method, and it has better robustness and accuracy.

Convolution neural network (CNN) has been know as a state-of-the-art technique for hyperspectral image (HSI) classification. However, the basic CNN architecture still has obstacles to deal with the intraclass diversity caused by spatial variability of spectral signature, especially in the case of highly limited training samples. Furthermore, they also have challenges to exploit the relationships between different features of the same object. In this paper, we propose a new network architecture based on matrix capsule network (MatCapNet) to alleviate these problems. To extract spectral-spatial features, the proposed network takes the 3-D neighboring blocks as input data. Specifically, the network consists of two conventional convolution layers, a primary matrix capsule layer, and a convolution matrix capsule layer. The proposed network uses spread loss to maximize the gap of activation values between the target class and other wrong classes. To learn more relevant instantiation parameters of the input data, a decoder network is used to reconstruct the input center spectral signature. Finally, the reconstruction loss and the spread loss are added at a certain ratio to optimize network parameters. Besides, in order to solve the problem of unbalanced sample sizes, the sample weight coefficient is utilized. The proposed approaches are carried out on two well-known HSI data sets. Compared with other state-of-the-art methods, the experiment results exhibit that the proposed network could provide a competitive advantage in terms of classification accuracy.

As part of investigating the human perception of polarized light, Haidinger’s brushes (HB) is well-known optical phenomena. Although differences in the human polarization perception were well known there was a limited quantifying to them. This paper presents a 3D Stokes model of the human polarized light perception, views the human perception of polarized light as a multivariable applied optics problem to simulation and explored the optical phenomena of Haidinger’s brushes. The images comparison verifies that the simulation patterns and detected images are consistent which proves the model is appropriate for simulating polarization perception of the human eye. Then this paper explored the optical limitations of perception and macula density. The combination of the 3D Stokes model and experimental verification opens up new possibilities to become an early diagnostic method for eye polarization sensitivity and macular degeneration. The objective is to give purpose and new mathematics understanding to the biomedical optical phenomenon.

The unsharp masking algorithm is an effective method for image spatial domain enhancement. The improved unsharp masking algorithm can achieve the sharpening effect of edge enhancement while suppressing noise. However, it is difficult to implement in real-time image processing systems due to the need to perform multiple spatial domain filtering and a large amount of calculation. In this paper, an improved unsharp masking image enhancement algorithm is implemented on FPGA. Compared with the software implementation, time optimization has been carried out to meet the requirements of real-time processing of video streams, and it can achieve edge enhancement while suppressing noise to meet the performance needs of real-time video processing.

The spatial resolution of hyperspectral data is low, so there are a large number of mixed pixels, which is also one of the main reasons that reduce the accuracy of hyperspectral image target classification. Hyperspectral unmixing is an important subject in the field of remote sensing. Hyperspectral unmixing generally consists of three steps: reduction, endmember extraction and inversion. As one of the key steps of hyperspectral unmixing, efficient and rapid endmember extraction is an important object in hyperspectral remote sensing. In this paper, the endmember extraction of hyperspectral data is implemented based on PCA and a new SGA algorithm, which solves the dimension limitation of traditional SGA algorithm and the new SGA algorithm without data redundancy caused by data dimensionality reduction.

The camera control system is the most important part of the space remote sensing camera. Due to the products in the camera of the Marine Satellite or the Meteorological Satellite are easily disturbed by the space environment, it is necessary to develop a high-reliability camera control technology to guarantee the camera reliable and long-life operation in orbit. Firstly, in this paper, a dynamic reconfiguration technology of a serial bus based on multi-node and extensible ways is carried out, and a fault detection and automatic recovery technology based on error detection and correction of hamming code and the periodic automatic refresh technology are applied. Meanwhile, the camera control system not only apply the single event immune memory to storage original program safely, but also carry out on-orbit maintenance technology which include parameter and program on-line reloading for the facilities in the whole camera, including control system itself. Secondly, In order to ensure the safety of infrared detection component, the control technology of early over-temperature warning and self-protection is also applied. Ultimately, The autonomous and interactive control technology of the multi-process imaging is developed for the multichannel camera, it can facilitate user to complete the imaging process to relay on ‘shortcuts’. Within conclusion, in this paper, The camera control technology realized can not only effectively guarantee the camera to cope with the harsh space environment, but also solve the convenience of users of the complex camera system, and it provides a technical guide for future development of space remote sensing camera control system.

Graphene-based photodetector with ultra-high responsivity is an important research field of low dimensional optoelectronics applications. A number of hybrid graphene/quantum dots photodetectors with high responsivity have been developed. In this paper, the in-situ oxidation of the copper covered by monolayer graphene was studied under the oxygen-rich condition. It is found that the oxidation process first occurs at the grain boundary of graphene and the oxide is Cu₂O. The intensity ratio of 2D band and G band of graphene is ~3, and the defect D peak is absent, which indicates that the quality of graphene is not damaged during the oxidation process. The hybrid transfer-free graphene/Cu₂O photodetector is fabricated by in-situ copper oxidation. Under 450 nm laser illumination, the responsivity of the photodetector is 3.8×10⁶ A/W at 0.2 V. The gain is up to 1.1×10⁷ , which is due to the modulation of Fermi level of graphene by Cu₂O quantum dots. The photodetector exhibits the specific detectivity of 3.6×10¹¹ Jones. This work opens a feasible pathway to develop transfer-free graphene/semiconductor photodetector with high responsivity.

We theoretically develop a method for maintaining the self-compensated state of a K-Rb²¹Ne comagnetometer, which is one of the core problems in optimizing the detection sensitivity of the comagnetometer to fundamental interactions and inertial rotation. The result of Rb²¹Ne coupling dynamics under transverse cosinusoidal magnetic field modulation can present a dispersive response with respect to the varying ²¹Ne nuclear magnetic field Bⁿ, in which a monotonic interval can be used to monitor the drift of Bⁿ. Furthermore, the result of Rb spin poparization evolution under the constraint of self-compensated-state maintenance exhibits a large enough time scale relative to Rb spin relaxation that it can be regarded as a quasi-static process, thus the adopted time-varying pump light intensity can be obtained using Rb equilibrium polarization. Accordingly, an implementation scheme is given. This method can not only compensate the drifting �", but also promote the system to return back to the original self-compensated state through manipulation of atomic polarization via time-varying pump light, which is useful in stabilizing the system performance.

With the continuous development of astronomy theory and space exploration technology, searching for extrasolar planets has become one of the most active research topics in astronomy. In recent decades, countries around the world have invested a lot of ground and space projects in this search field and obtained abundant results. Firstly, this paper summarizes the mainstream exoplanet detection methods such as radial velocity, transit and direct imaging method with the outline of the principles and features. Then, several instruments for obtaining the spectrum of exoplanets are introduced, focusing on the optical system parameters of telescopes and spectrometers. Finally, according to the comprehensive discussion above, the future development trend of exploration missions and instrument design in this field is predicted, and it is recognized that these survey missions for detecting and characterizing exoplanets are of great significance for searching biological signals outside the solar system.

In this paper, the photometric characteristics of passively marked corner-cube-reflector-class cooperative targets are studied. The imaging characteristics of corner cube reflector(CCR) are discussed theoretically, and the influencing factors such as surface accuracy, angle error and diffraction effect are analyzed. The reflection uniformity and diffraction characteristics of CCRs with different positions and different precision are simulated and verified. In addition, the influencing factors and control methods of CCR precision are proposed. To the photometric characteristics of cooperative targets, especially the influence of aperture size of cube-corner prism on comprehensive aberration, the transmission surface (bottom surface) on comprehensive aberration was concerned firstly, besides the three reflection surfaces. The conclusion was been drawn that the transmission surface and the three reflection surfaces had most effect on the parallel of emerging beams, and the minimum error was the minimum algebra sum of above four surfaces. It can provide theoretical support for on-orbit services such as companion flight, autonomous rendezvous and docking (RVD).

In the process of surface reconstruction, the gradient data inevitably contains noise, which results in poor surface quality. Traditional denoising algorithms have slow processing speed and unsatisfied effects. To realize high precision surface reconstruction, a novel denoising method was presented based on improved curvature filter. The innovative method optimizes the projection operator to enhance the denoising capability of the algorithm. According to the numerical simulation and analysis, the simulation results show that the proposed algorithm can remove noise in the gradient data and contributes to higher accuracy of surface reconstruction.

Tunable diode laser absorption spectroscopy (TDLAS) technique has been widely investigated for gas concentration measurement in both industry and laboratory. In order to detect different gases within the multi-gas mixture based on TDLAS, different types of schemes have been developed, such as wavelength division multiplexing, time division multiplexing and so on. However, there are many drawbacks of the above methods, and the sensitivity and accuracy of multi-gas detection have been greatly restricted, the effect of cross-talking interferences becomes a technical bottleneck for multi-gas detection. Therefore, a high accuracy synchronous detection technology for multi-gas detection using the least square fitting is reported. The wavenumber of 6380cm^-1 has been selected to detect CO and CH₄. Because the absorptivity of CH₄ and CO is less than 0.1, the least square fitting method can be used to calculate the concentrations of CH₄ and CO simultaneously. This novel method has been shown to improve the precision achieved in the detection of multi-gas by 18%, compared with the precision measured at another wavelength. The Allan variance results indicate that the optimal integration time has been improved from 50s to 100s, the minimum measurement precision of CH₄ and CO is ~0.45% and ~0.46 10-6 respectively. Meanwhile, the detection cost and response time can be reduced obviously.

This paper presents a mathematical model which can fit a parameter in an optimal position by measuring data. The model can synchronously correct the measurement error caused by the small rotation about X, Y and Z axes and the small deviation along X, Y and Z directions. The measurement results of the surface show that the model is reliable and effective, which provides loose conditions for placing mirrors and reliable measurement results for contact Coordinate Measuring Machine(CMM) off-axis surface contour measurement. The basic principle of the data analysis is also of high reference value for contour measurement of other non-contact free-form surfaces. The validity of the model is verified by mathematical simulation and practical engineering application, which is of great help to the measurement of high precision contact free-form surface.

Compared with the image target detection system of single infrared band , the target detection system of double infrared band image fusion has higher performance. However, the target detection of dual band image fusion requires higher real-time and stability of data processing. Aiming at the characteristics of large amount of data and complex algorithm of dual band image fusion target detection system, a dual band infrared image target detection system based on DSP+FPGA architecture is designed. This system has the advantage that DSP is good at realizing complex image processing algorithms and FPGA has high speed parallel processing capacity. This design meets the system's requirements for real time and stability.

Fiber optic imaging elements is a general term for optical fiber products including fiber optic plate (FOP), fiber optic inverter (FOI), and fiber optic taper (FOT), which is a light and image transmission array made of optical glass fibers that are drawn, regularly arranged, vacuum melted-pressed, secondary thermal processed and optically processed. They are mainly used in photomultiplier devices for low-light-level night vision and particle detection equipment. Its devices have been widely used in many fields such as low-light night vision, electronics, aerospace, nuclear diagnostics, high-speed photography, etc., especially in the field of military operations at night combat, guidance, early warning, and electro-optical countermeasures. With the continuous development of glass fiber material, preparation technology and performance requirements, the optical fiber imaging element has been significantly improved, and its application field has been continuously expanded. This paper reviews the materials, structures, principles, characteristics, preparation processes and detection technologies of optic fiber imaging elements. It mainly summarizes the development situation, latest technological progress, and several major devices in optical applications. And the bottlenecks in the preparation and detection of fiber-optic imaging elements at present are analyzed. Furthermore, the development trend and challenges of fiber-optic imaging elements evolving to larger size with smaller diameter of fiber, higher contrast, higher resolution, lower distortion, ultra-short and ultra-light are prospected.

High-power special-shaped infrared radiation sources are usually used for detection systems such as compact photoacoustic spectrometers for gas detection. In this article, the light emitting characteristics of a rod-shaped infrared thermal radiation light source was obtained experimentally. Then, based on the luminous characteristics and the illuminance requirements on the target photoacoustic cell, combined with the basic principle of conservation of light energy during transmission, the shaping system for the special-shaped infrared radiation source for photoacoustic spectroscopy was designed and optimized, aiming to achieve efficient collection and utilization of light source beams.

The analog signal output by the infrared detector is usually sampled as a 14-bit high-dynamic-range digital infrared image, which is not compatible with most 8-bit display systems and cannot be directly used for imaging display. Therefore, image compression quality is very important for image display .This paper studies and analyzes the mapping technology of the base layer, focusing on the traditional limited contrast histogram equalization method (CLAHE). Considering that the conventional CLAHE method is applied to infrared images, there will be a problem of excessive stretching of some uniform areas. a new algorithm for Contrasted Limited Adaptive Histogram Equalization based on uniform region test (M-CLAHE) is proposed. Based on the traditional CLAHE method, this method introduces a gradient operator to extract the contour boundary of the sub-region image, and quantifies the complexity of the image region to solve the above problems. The experimental results show that compared with the traditional CLAHE infrared image compression method, this method can more than double the visual indicators such as image contrast and average gradient compared with the traditional visualization method. This method has engineering portability, and it can be transplanted to the infrared movement to increase the image quality of most scenes by about twice.

Space cryogenic infrared remote sensor has a wide range of applications, but there are only a few cases which function well in orbits, especially the domestic related research is still in the preliminary stage. Cryogenic lens is installed and adjusted at room temperature. In low temperature environment, the lens structure will produce thermal stress deformation, which will affect the surface shape and focal plane position of the lens, thus affecting the optical imaging quality. Therefore, the design of cryogenic lens must combine the three physical fields of light, machine and heat. This article is based on an infrared cryogenic lens, low temperature 210k working conditions, cryogenic lens finite element model (FEM) is established, using the simulation analysis software to calculate the lens at low temperature for the thermal deformation and thermal stress, the deformation of the mirror node data export for Zernike polynomial fitting, and translated into changes in surface shape and deformation after the modulation transfer function (MTF). After the installation and adjustment, the cryogenic lens was placed in a vacuum tank, and MTF was tested at room temperature and low temperature. The test results showed that the deviation is less than 0.5%, within the acceptable range, indicating that the cryogenic lens multi-field coupling simulation method used in this paper is reliable and has strong engineering significance.

In order to obtain reflector supporting structure with high stiffness and thermal stability, a “2-2-1-1” kinematic supporting structure is designed for a 350mm mirror of space remote sensor. Firstly, the kinematic principle of the “2-2-1-1” supporting structure is studied, and the dynamic model of the “2-2-1-1” supporting structure is deduced, the relationship between the supporting stiffness of the mirror assembly and the parameters of the length, diameter and span of the ball-end supporting rod is given. According to the relationship between the supporting stiffness of the mirror assembly and the length and diameter of the supporting rod, a “2-2-1-1” supporting structure of the mirror assembly is designed. The mechanical simulation of the design results is carried out by means of finite element analysis, and then the test of the actual supporting structure is carried out. The experimental results show that the RMS value of the reflector supported by “2-2-1-1” supporting structure is better than λ/40(λ=632.8nm), The fundamental frequencies of the components in three directions are respectively 231Hz, 319Hz and 365Hz, it is close to the results of finite element analysis and theoretical calculation. The rationality of the designing of the “2-2-1-1” supporting structure of the reflector is proved, which meets the imaging requirements of the space remote sensor.

The photoelectric tracking platform is often installed in various military and civilian equipment such as unmanned aerial vehicles, and has multiple high-precision detection and tracking equipment such as visible light and infrared thermal imaging cameras. In the detection process of various detection imaging systems, because the optical detection equipment is fixed on a carrier aircraft such as an unmanned aerial vehicle, the imaging system often rotates the detection image due to the rolling, pitch and yaw movement of itself and the carrier. In this regard, this article will analyze the principle of the image rotation problem caused by the photoelectric tracking platform in the detection and tracking of the target, and design an algorithm to eliminate the image rotation. First of all, this article clearly understands the purpose and indicators of rotation, carefully analyzes the reasons for the problem of image rotation, improves the definition of the relevant coordinate system and the relationship between them, and finally derives the value of the image rotation angle and performs the operation of eliminating the image rotation. And image post-processing operations. After testing, the designed algorithm can solve the problem of rotation of the image presented by the photoelectric tracking platform, and the processed image is easier to observe and capture manually or by related algorithms.

Direct detection laser detection and ranging (LADAR) has been widely used in many specific applications such as precision guidance, machine vision, underwater images, landslides investigations, city modeling, and so forth. One of the most promising methods to design, develop, test and validate a LADAR system is hardware-in-the-loop (HWIL) simulation. LADAR target simulator generates return signals of the simulated targets and background according to the test requirements. One of the key technologies of the LADAR target simulator is target modeling. A target modeling method based on data matching and restoring is pdroposed in this paper, which can provide a feasible way for LADAR target simulator to acquire the information of the target. The coordinate transformation is used to transform the target coordinate data under two or more viewpoint coordinate system into the reference point coordinate system to obtain the whole information of the target. Then, the target coordinate data under the reference coordinate system is transformed into a given viewpoint coordinate system. Furthermore, the deley time and the pulse width are calculated arrcording to a return signal mathematical model. Finally, the effects of the incident angle and target distance on target data are analyzed.

In order to clarify the mechanism on electron multiplication of MCP, it is necessary to study the secondary electron emission properties of inner wall surface. The secondary electron emission is not completely required by materials or devices, and sometimes it can bring bad effects such as noise. In addition to the secondary electron produced by electron colliding with the inner wall, there are additional secondary electrons produced by the ionized cation colliding with the inner wall, which is the key to control the noise of MCP. In this paper, the secondary electron emission produced by the ionized cation colliding with the inner wall of MCP was analyzed after the materials were acid and alkaline treated and hydrogen reduced at high temperature. Time of Flying (ToF) was used to detect secondary electron emission yield (SEEY) of lead silicate glass surface. By comparative analyses, the influence trend of surface treatment process on the secondary electron emission yield of glass surface was acquired. It is expected to provide data supports for parameters optimization of MCP manufacture and preparation of MCP with high signal to noise ratio (SNR).

Three-dimensional measurement technology is a popular technology in recent years. The traditional binocular vision measurement system has a low image matching accuracy when there are few or many duplicate feature points on the surface of the object to be measured, which affects the work of three-dimensional reconstruction. In this paper, the combination of structured light three-dimensional measurement technology and binocular imaging technology reduces the difficulty of feature point search, increases the accuracy of image matching, and conducts experiments of three-dimensional measurement by building a verification system. The measurement results show that the technology is highly feasible.

Scene classification shows pivotal role in remote sensing image researches. Since challenges of large similarity between classes, high diversity in each class and huge variations in background, spatial resolution, translation, etc., remote sensing image scene classification still urgently need development. In this paper, we propose a novel method named deep combinative feature learning (DCFL) to extract low-level texture and high-level semantic information from different network layers. First, feature encoder VGGNet-16 is fine-tuned for subsequent multi-scale feature extraction. And two shallow convolutional (Conv) layers are selected for convolutional feature summing maps (CFSM), from which we extract uniform LBP with rotation invariance to excavate detailed texture. Deep semantic features from fully-connected (FC) layer concatenated with shallow detailed features constitute deep combinative features, which are thrown into support vector machine (SVM) classifier for final classification. Extensive experiments are carried out and results prove the comparable advantages and effectiveness of the proposed DCFL contrasting with different state-of-art methods.

In order to realize the in situ detection of plankton and organic particles in the deep sea, we proposed a dynamic holographic microscopic imaging method based on the principle of parallel phase-shifting. We analyzed the implementation mechanism of the parallel phase-shifting holographic imaging. We designed the implementation scheme of the deep sea in situ detection system, carrying out experimental research on underwater dynamic holography. High resolution images were acquired and reconstructed using the resolution target and several biological samples. Experimental results show that local details of underwater samples can be clearly distinguished. We preliminarily verified the feasibility of the parallel phaseshifting holographic method to achieve underwater high-resolution in situ detection.

A micro filter array for portable spectrometer was fabricated. Based on the Fabry Perot periodic multilayer dielectric film, a narrow band filter with a central wavelength of 589 nm was designed and fabricated. The bandwidth is about 3.2 nm, the transmittance is more than 60%, and the cutoff is OD3 . The mask is designed based on the selected mobile camera, and the pixel of narrow-band filter is realized by using semiconductor technology such as step lithography and reactive ion etching on the optical film. The optical fiber spectrometer and atomic force microscope (AFM) were used to measure the micro area of the filter array. The results show that the adjustment of the Fabry Perot cavity is accurate and effective. A 4 × 4 filter periodic structure is realized. The filter array with unit size of 3 μm × 3 μm and size of 1280 × 720 is fabricated.

The surfaces of many ancient relics as murals and oil paintings we can see today appear three-dimensional features to some extent, which poses a challenging problem for line scan imaging to capture the features on these objects. With this goal in mind, a line scan camera-based shape from focus (SFF) method is proposed for obtaining three-dimensional information in planar cultural heritages with three-dimensional features. First, line scan imaging is employed to obtain high-resolution image sequence of objects. Next, spatial frequency (SF) is used to evaluate the focus quality of a specific local area of each image in the image sequence to find out the position of the best focused image. Through analyzing the relationship between this position and the relative movement distance of the camera-object, the three-dimensional shape of the object can be roughly estimated. Then, a fitting method was utilized to refine the local details of the three- dimensional shape of the object. An experimental setup is established to restore the three-dimensional shape of the test sample using the proposed SFF algorithm. Root mean square error (RMSE) was used as the evaluation criteria, and the result demonstrated the effectiveness of the proposed method, which shows the high precision three-dimensional shape recovery and also gains the benefits of high resolution and uniform point cloud distribution. Finally, the factors affecting the results of 3D reconstruction was discussed in detail, which may guide the selection of parameters in practice.

With the development of the fourier ptychography in microscopy imaging, more and more researchers consider applying fourier ptychography in long-distance imaging. However being different from microscopic imaging, long distance fourier ptychographic imaging will face more challenges, one of which is the source image quality. In this manuscript, the influence of the source image quality on camera scanning FP imaging will be investigated and simulated. In the first part, the calibration aims to solve the influence of unenenly illumination. After that, a poisson-gaussian mixed noise model based denoising is uesd and could effectively suppresses the noise through parameter estimation. Finally, Phase correalation regisitrtion has to be used to correct the mismatch between adjacent images caused by camera scanning mode. The simulation results demonstrate the effectiveness of the preprocessing methods and could make FP more robust.The signal to noise ratio could reach to 40.5dB while obtained 5×improvement in resolution.

The SPGD (stochastic parallel gradient descent) is a control algorithm widely used in WFSless (wavefront sensorless) AO (adaptive optics) system.The gain is commonly set to a fixed value in the traditional SPGD algorithm.With the increase of the number of DM (deformable mirror) actuators, the optimization space of the algorithm becomes larger, which can easily lead to the slow convergence speed of the algorithm and the rise of the probability of falling into the local optima. Adam(Adaptive Moment Estimation) optimizer is an optimized stochastic gradient descent algorithm commonly used in deep learning with the advantage of achieving adaptive gain. Wavefront aberrations under different turbulence strength as correction objects, WFSless AO systems are built with 32, 61, 97 and 127 elements DM as wavefront corrector respectively. Results show that the optimized algorithm can converge faster than basic SPGD and the probability of falling into local optima decreases. The system’s convergence speed is increased by about 30%. The probability of falling into local optima is decreased by 29.8% , 30.3% , 32.6% and 35.9% respectively under D/r0=5 and by 28.8% , 30.5% , 33.3% and 34.5% respectively under D/r₀=15. The advantages of the optimized algorithm are more obvious as the number of DM actuators increases. Above results provide a theoretical basis for the practical application of the SPGD algorithm based on Adam optimization.

The full-stokes expressions of polarizer direction (PD) and Quarter-wave-plate direction (QWD) was deduced at any angle. The experiments of full-stokes polarization imaging detection were carried out with artificial target under the natural background, and the degree of polarization (DOP) images contrast (DOPIC) with the target and the background were analyzed. This paper mainly research that different angle combination influence between polarizer and quarter-wave plate for the DOPIC of full-stokes polarization imaging detection. It was found that the DOPIC of target and background was obviously higher than intensity image, but also (1)The detection results were related to the angle of PD when changing the angle of QWD only; (2) If three groups of direction angles (α₁β₁α₂β₂α₃β₃) were same and the fourth group of angles (α₄β₄) was set interval angle 45°between PD and QWD angle, the average detection result of full-stokes polarization imaging was higher than other PD and QWD.

High-precision phase retrieval is the key step of the fringe projection profilometry, while the traditional multifrequency phase-shifting method is easy to suffer from noise, which leads to a decrease in measurement accuracy. Based on the local similarity of phases, a new phase retrieval method is proposed in order to improve the accuracy of the phaseshifting and reduce the phase order error. The input images are down-sampled to obtain an image pyramid, and the phase of the high-resolution images is recovered from the low-resolution fringe patterns. Simulation and experimental results show that the proposed method can effectively reduce the phase error in the multi-frequency phase-shifting method and has good robustness and high accuracy

Position sensitive micro-channel plate (MCP ) is a kind of high sensitivity detector with low noise, Combined with advanced time-correlated single photon imaging technique (TCSPC) could make a moderate aperture system be sensitive to only one single arriving photon. In this paper, a theoretical model of photon counting imaging with Position sensitive MCP was established and the factor of laser power affecting the detect possibility was analyzed. the three-dimensional point clouds data are generated using Monte Carlo simulation and the typical space target could be reconstructed three-dimensionally. Through the all-chain simulation model the comprehensive performance of the MCP based active three-dimensional imaging system could be analyzed from viewpoint of detection probability, ranging accuracy and signal to noise ratio etc. The application of the position sensitive MCP based active three-dimensional imaging system for long range space objects is verified in simulation condition. The results shows that only a single pulse energy of tens of mico-Joule is needed for the positive sensitive MCP based active three-dimensional imaging system to image the target at the hundreds of kilometers. the counting rate could reach 106 counts/s. And the ranging accuracy of this active three-dimensional imaging system for the objects at 300 kilometers could reach 0.1228meters by simulating with 50 mico-Joule single pulse power.

Linear array CCD device has the characteristics of large pixel size, high precision, good resolution, wide spectral response, small volume and so on. It is widely used in image sensing, spectral analysis and non-contact measurement fields such as size, displacement and vibration measurement.In this paper, the general structure and timing control characteristics of linear CCD are analyzed, and the typical CCD chip Toshiba TCD1304 is studied. The CCD is a linear image sensor with high sensitivity and low dark current. The pixel size is 8um × 200um, 3648 pixels in total, and the wavelength response range is 300nm ~ 1100nm.Secondly, the 32-bit flash microcontroller STM32F103 based on the core of arm cortex m3 processor is used as the main control chip, and a set of CCD data acquisition system with flexible integration time adjustment function is designed by combining with the 12 bit high speed A/D converter, 2.8-inch TFT LCD and upper computer software.Finally, the whole system is tested, The test results of the system verify the feasibility of the driving timing design of linear array CCD. The driving timing signal is connected to the CCD device. Under the condition of different light incidence, the CCD works normally and outputs the corresponding video signal driven by the driving signal. The technical indexes of the system can meet the requirements of the system design, and provide a technical reference for similar CCD data acquisition system and lay the foundation for further image recognition and processing.

Photoelectric stabilization platform is the eye of the aircraft.At present, the two-axis two-gimbal photoelectric platform is widely used in military, commercial and scientific applications due to simple mechanism, low cost and small size. However, the stability performance of the outer gimbal will be reduced at large pitching angle. In this paper the reason was analyzed , the three-axis gyro system was proposed to stabilize the outer gimbal and the mechanism of this method was explained ,then the control system was designed, the outputs of speed loop were compared between the three-axis gyro and the traditional two-axis gyro system, The results showed three-axis gyro system has higher stabilization accuracy than two-axis gyro system.

The imaging principle and radiometric calibration principle of long linear array TDICCD cameras are analyzed. Calibration work is carried out using the integrating sphere and collimator while developing a more comprehensive radiometric calibration scheme. The study used quantitative analysis to gain insights into the linear responsivity of input radiance, integral time, integral series, and gain to the camera. The least-square method is chosen to fit the experimental data to obtain the corresponding linear regression equation. The correlation coefficient is above 0.99. The results show that the radiation response of TDICCD camera is excellent when other dimming parameters remain constant; the gray value of the image changes linearly with the input radiance value. Fixing the input radiance, the gray value of the image varies linearly with the gain, integral series, or integral time. The results of this research support the study of the automatic dimming of TDICCD camera.