To test whether the fiber Bragg grating (FBG) sensor can endure the steady-state inertial loads caused by the acceleration and the sensing properties during the loads, a FBG strain and temperature sensor with aluminium alloy substrate package was designed, and the acceleration performance on the sensor was tested. The sizes of FBG strain and temperature sensor were designed and its package process was described. The strain and temperature sensing mechanisms of FBG sensor were analyzed, and the spectrum detection and demodulation system based on volume phase grating and linear array photodetector was developed. Finally, the acceleration test equipment was established, and the acceleration performance test of the selected FBG strain and temperature sensor was carried out in accordance with the requirements and methods of GJB150.15A acceleration test. The experimental results show that in the 2 min performance test before and after the acceleration test, the wavelength offset is below to ±50 pm, and the change of light intensity is below to 0.3 V. In acceleration test, the maximum fluctuation of wavelength offset is ±7 pm, and the light intensity is in the range of 1.3 V～4.003 V. It is proved that the designed FBG sensor has the ability to endure the acceleration loads and has the good sensing performance during the acceleration loads.

The demodulation method of fiber Bragg gratings (FBG) based on long-period fiber gratings (LPFG) was studied. The transmission spectrum of LPFG was used as the edge filter to demodulate the FBG, and the demodulation performance mainly depended on the transmission spectrum depth of LPFG. The influence of CO2 laser writing technique on the transmission spectrum was analyzed in detail, and the LPFG with transmission spectrum depth of 25 dB was manufactured and used as the FBG demodulation device. The demodulation system was built with the FBG pasted on the surface of the metal strain sheet. The static and dynamic response of the sensing system was tested. The experimental results show that the proposed method has the advantages of simple production process, low cost, good linear dynamic range and higher sensitivity.

A temperature sensor based on coreless-few mode-coreless optical fiber structure was proposed for theoretical analysis and experimental study. The coreless fiber (CLF) and the few-mode fiber (FMF) were fused together to form a coreless-few mode-coreless optical fiber structure, and the single-mode fiber (SMF) was fused at both ends of the structure as input and output fiber. The mode mismatch between the first section of coreless fiber and single-mode fiber could excite higher-order modes. The two modes of LP01 and LP11 in the few-mode fiber were transmitted along the core of the few-mode fiber. Under the action of the second section coreless fiber, the two modes were recoupled back to the single-mode fiber, and the two modes interfered to form an interference spectrum. When the outside temperature changed, the optical path difference between the two modes also changed, and the interference troughs of the interference spectrum were shifted. Two different interference troughs were selected as the characteristic wavelengths for experimental analysis. The experimental results show that the interference troughs with wavelength around 1 550 nm and 1 534 nm both have red shift, and the corresponding temperature sensitivity is 68 pm/ and 44.5 pm/ respectively. The sensing structure has the advantages of simple fabrication, high sensitivity and good application prospects.

In order to meet the demand of microseismic monitoring in oil and gas development, a new type of underground three components optical fiber microseismic monitoring system was designed based on time-division multiplexing scheme. Combined with indoor test, the background noise of the system was less than -101 dB, dynamic range more than 120 dB, interstage crosstalk less than -67 dB, which meet the needs of microseismic signal detection. The system was successfully applied to the field detection of hydraulic fracturing microseism in Xinjiang Oilfield. The monitoring results show that the system can clearly capture the P-wave and S-wave signals of microseism, better reveal the characteristic information of hydraulic fracturing microseismic signals, and establish a good foundation for the spatial location of microseismic events and the interpretation of fracture characteristic parameters.

High precision grating diffraction laser warning system is a photoelectric countermeasure equipment to judge and locate the incoming laser function. In order to improve the positioning accuracy, an improved algorithm for calculating and fitting the calculation parameters of the spot center was proposed. Firstly, the laser spot was preprocessed, and the center coordinates of the connected area were obtained by using the improved connected area marking algorithm and spot center extraction algorithm. Secondly, the azimuth angle and pitch angle were set, the images were collected and the spot center was calculated, and the pixel horizontal and vertical coordinates in x and y direction as well as fitting surfaces of laser parameters were fitted, so as to judge the azimuth angle, pitch angle and wavelength of the incoming laser. Finally, the improved connected area marking algorithm, spot center extraction algorithm and parameter fitting calculation results were compared with the traditional algorithm. The experimental results show that the pitch angle error between the fitting calculation results and the real data is less than 0.4°, and the azimuth angle error is less than 0.2°, which greatly improves the accuracy of laser direction recognition.

Based on the basic theory of lightwave diffraction, the regulation method of beam distortion in the diffraction process of the Kepler telescope system was studied by using a Gaussian-distributed laser beam. The aberration function method was constructed by using the intermediate-order Zernike polynomial to simulate the four-leaf distortion in the telescope system, and the Zernike factor was introduced into the optical system to compensate for the distortion effect. The results show that the wavelength and aperture have a significant influence on the distribution of singular light intensity, in which the increase of wavelength will change the shape of the light spot, the aperture will affect the intensity and size of the light spot, and the 11th order Zernike four-leaf deformation factor can offset the distortion phenomenon of the telescope system.

The ghost imaging is an imaging technology that can penetrate harsh environments such as the heavy fog. Aiming at the problems of more noise and lower image contrast of reconstructed images of traditional ghost imaging, the non-local generalized total variation method was applied for image reconstruction of ghost imaging, and the reconstruction method of computational ghost imaging based on non-local generalized total variation was proposed. The method constructed the non-local correlation weights to design the gradient operator, which was substituted into total variation reconstruction algorithm, so that the reconstructed images could effectively remove the noise while achieving the better detail restoration. The simulations were performed under different conditions, and the peak signal-to-noise ratio of proposed method was improved by about 1 dB compared with other methods, while it had better subjective visual effects. The experimental platform was designed and built to verify the effectiveness of the algorithm. The experimental results verify the superiority of the proposed method in terms of noise removal and detail reconstruction.

Aiming at the problems of low efficiency and low accuracy in the detection of internal assembly defects of thermal batteries, a method which could accurately segment the internal battery stack images and accurately identify the types of defects was studied. Firstly, the horizontal and vertical integral projection methods were used to extract the edge features of the target battery stack, and the local adaptive contrast enhancement algorithm was used to enhance the detail texture of the local unclear parts. Then, the gray characteristics of the defect structure were studied and the defect characteristic parameters were calculated and extracted. Finally, the BP neural network and CART decision tree were used to classify and identify the feature parameters, the weight was allocated according to the classification accuracy, and the weighted fusion results were used as the final criterion of the detection. The experimental results show that the accuracy of this method is 98.9% for 2 000 samples, which provides an effective way for X-ray defects detection of thermal batteries.

Aiming at the problems of mutual occlusion between assembly parts, different poses of parts, external light intensity, missed detection of small targets and low detection accuracy of traditional machine vision detection and recognition methods, a parts recognition method based on improved faster recurrent convolutional neural network (RCNN) was proposed. Firstly, the ResNet101 network with better feature extraction was used to replace VGG16 feature extraction network in original Faster RCNN model. Secondly, for the original candidate region network, the two new anchors were added and the aspect ratio of candidate frame was reset to obtain the 15 anchors with different sizes. Then, aiming at the missed detection problems caused by deleting the candidate frame in which the Intersection-over-Union (IoU) was greater than the threshold in traditional non-maximum suppression (NMS) method, the Soft-NMS method was used to replace the traditional NMS method, so as to reduce the missed detection problems in dense regions. Finally, in training model stage, the multi-scale training strategy was adopted to reduce the missed detection rate and improve the accuracy of the model. The experimental results show that the improved Faster RCNN model can achieve 96.1% accuracy, which is 4.6% higher than the original model, and can meet the recognition and detection of parts in complex conditions such as strong illumination and water stain interference.

The atmospheric turbulence effect is one of the important factors that seriously affect the image quality of aerial photoelectric reconnaissance system. Starting from the description of atmospheric turbulence parameters, the influence mechanism of atmospheric turbulence on the imaging quality of optical system was researched, the influencing factors of atmospheric turbulence on the modulation transfer function (MTF) of optical system were analyzed, and the theoretical model of atmospheric turbulence on the imaging MTF of optical system was established. The simulation results show that under the influence of atmospheric turbulence, the ratio of optical system pupil aperture to atmospheric coherence diameter has a great influence on the imaging MTF. By comparing the ground field experimental images of the optical system, the imaging MTF theoretical model of the actual optical system affected by atmospheric turbulence was verified. The research results can provide theoretical supports for the design of aerial photoelectric reconnaissance system and the improvement of imaging quality under the influence of atmospheric turbulence.

The imaging luminance measurement device (ILMD) can utilize the image sensor and the short focal length lens to achieve a large field of view and spatially resolved luminance measurement. However, there still exists problems of pixel nonlinear response of image sensor, strong vignetting effects of short focal length lens and image distortion. Therefore, a correction method of ILMD was proposed. Linear correction and flat-field correction were used to obtain linear correction coefficients and flat-field correction matrices by using standard radiation source method. The distortion correction matrix was obtained through the geometric coordinates calibration method. A 12 mm short focal length lens and a 2-megapixel image sensor were adopted to build an ILMD. After correction, the luminance measurement of the LCD panel was completed. It was compared with a commercial spectroradiometer and the test results show that the relative error of measurement does not exceed ±2%, which realizes the large field of view, high-precision and spatially resolved luminance measurement.

Aiming at the detection requirements of helicopter networking and informatization, a method of using data link information to quickly guide the airborne electro-optical system realizing target aiming was proposed. The indication errors during the guidance process of network information were analyzed. The coordinate transformation method and delay processing algorithm of target information, and the aiming line deviation estimation method of electro-optical system were given. Finally, the effects of data accuracy, data rate, delay time on network cooperative target aiming for helicopter electro-optical system were studied by simulation experiment. The simulation results show that reducing the data latency, improving the data rate and data accuracy are the effective approaches of reducing the aiming error of network cooperative target. In addition, it can improve the estimation accuracy of target state and reduce the estimation errors of target aiming line by using the reasonable target information delay processing algorithm.

In order to improve the control degree of freedom and surface shape accuracy of the piezoelectric bimorph mirror (PBM) in synchrotron radiation, and to solve the problem of abnormal fluctuations of the calculated voltages (overfitting) affected by noise caused by the excessive number of piezoelectric actuation units, the PBM model was established and simulation control was carried out. The 36 sets of piezoelectric response equations were obtained through the finite element simulation, and the mathematical model of surface shape and voltage was constructed. To compensate the mirror surface distortion caused by gravity, the control effects of two voltage solutions using least square method (LSM) and Tikhonov regularization were analyzed and compared by obtained ellipse shape. The results show that after using Tikhonov regularization inversion, the surface shape control error is reduced by 21.7% compared with LSM, and the maximum voltage fluctuation between adjacent electrodes is reduced from 1.019 kV to 0.174 kV, which meet the actual requirements of engineering. The system is robust to tested noise, and has more superior application value than LSM.

In order to realize the image quality detection and evaluation of photoelectric imaging system under the condition of temperature change, an optical window with temperature adaptation function was designed. Firstly, the influence of temperature on the surface shape of optical glass was analyzed. Secondly, the temperature adaptability opto-mechanical structure of optical window was designed. Finally, the effect of temperature change on the surface shape of optical window was analyzed by means of finite element analysis and experiment, and the validity of temperature adaptability design was verified. The experimental results show that the peak-valley (PV) and root-mean-square (RMS) values of the wave aberration of optical window are 82.90 nm and 6.96 nm respectively at room temperature of 20 ℃, the PV and RMS values of the wave aberration are 136.68 nm and 14.55 nm respectively at high temperature of 50 ℃, and the PV and RMS values of the wave aberration are 183.51 nm and 28.48 nm respectively at low temperature of -40 ℃. Under high and low temperature conditions and compared with the room temperature condition, the numerical variation trend of wave aberration is in good agreement with the finite element analysis results. Under three temperature conditions, the PV values of the wave aberration of optical window are less than or close to (1/4) λ, and the change of the surface shape of optical window caused by the temperature change is very small. The designed optical window has better temperature adaptability.

There exist friction torque, unbalanced torque and other disturbance torques in the process of image stabilization and target tracking by photoelectric pod, which affect the response accuracy of speed loop. On the other hand, the delay caused by the image transmission and processing of the pod video tracker can cause tracking lag. Therefore, the delay must be compensated so as to reduce the error. For this, a sliding mode variable structure control method based on prediction tracking was proposed, in which the differential prediction tracker was adopted to compensate the delay of video tracker, and the motion angular velocity of target estimated by the prediction tracker was applied to compose the adaptive parameter so as to adjust the quantity of sliding mode variable structure contro. The improved sliding mode control algorithm could compensate the disturbance torque and suppress the chattering phenomenon at the same time. The simulation and experiment results show that the improved control method can effectively compensate the errors caused by disturbance torque and delay of tracker , its tracking error is reduced to 1/3 of the original traditional PID control.What's more, the proposed method has been applied in related systems.

The application of infrared stealth technology in the stealth missile could reduce the detection probability of missile by infrared detection system. Taking AGM-158C missile as a reference, the infrared radiation characteristics of target in the head-on observation direction including target skin, target skin reflection of surrounding backgrounds and infrared radiation intensity of target tail flame were estimated, and were compared with the infrared radiation characteristics of typical subsonic missile. The analysis results show that the infrared stealth technology can significantly reduce the medium and long wave infrared radiation of the missile, especially in the medium wave band. In the design of the infrared detection system, the medium-wave infrared detector is usually used for the tail flame detection, while the long-wave infrared detector used for the missile body detection. The stealth missile using the infrared stealth technology will improve the penetration ability of the missile.

A wavefront evaluation method based on sparse subaperture was proposed for the collimated wavefront of 300 mm aperture wavelength-tuned interferometer. The method used the wavefront data of sparse aperture to construct a uniform and equally spaced subaperture arrangement model and utilized the simultaneous fitting algorithm to realize the reconstruction of the full-aperture collimated wavefront. The change rules of subaperture spacing and subaperture size on the reconstruction accuracy were analyzed by the numerical calculation, and the optimized subaperture arrangement way was obtained. Finally, an optimized subaperture arrangement with a subaperture size of 10.8 mm and an adjacent subaperture center spacing of 9.72 mm was used for the sparse subaperture evaluation of 300 mm aperture collimated wavefront. The simulation results show that the optimized sparse subaperture evaluation wavefront residual peak valley (PV) value is 0.001 6 ${\rm{\lambda}}$ and the residual root mean square (RMS) value is 1.689 3e-4 ${\rm{\lambda}}$.

In order to solve the problems of low filling rate and easy deformation of imprinting patterns in the process of nanoimprint, a new vibration-assisted nanoimprint method was proposed. In the process of imprinting, the transverse vibration was applied to the imprinting adhesive, which increased the imprinting force and improved the filling rate. Using finite difference time domain (FDTD) method, different grating structures were numerically simulated in the wavelength range of 300 nm~1000 nm, and the influence rule of grating structure parameters on its absorptivity was obtained. Finally, the vibration-assisted nanoimprint experiment was carried out on the vibration-assisted device. The experimental results show that compared with the traditional nanoimprint technology, the filling rate of imprinting adhesive is increased by 30%, the surface morphology of microstructure after imprinting is improved and the defects are reduced.

The Ritchey-Chretien (R-C) optical system, as a two-mirror reflective optical system, is the typical structural form of hyperboloid optical system and has important applications in modern optical engineering. To shorten the adjustment cycle and guide the mass production of high-precision system, taking the R-C optical system applied in a certain airborne star sensor as an example, a secondary-mirror adjustment mechanism was designed for the secondary-mirror with small diameter under the premise of taking the primary mirror as the adjustment reference, which realized the adjustment of six degrees of freedom of secondary-mirror, and explained the method involved in the adjustment and curing process. The stability of the design mechanism was verified by testing the system image quality before and after adjustment and environmental test. The test results show that the secondary-mirror adjustment mechanism can meet the requirements of system image quality, the whole-machine wavefront aberration PV value is better than (1/2)λ, the RMS is better than (1/15)λ, and the adjustment method attains the goal of guiding the mass production of high-precision system.

In the process of using LED as lighting source, the optical field is the key factor in LED applications and is regulated by secondary optical system, which is complex in design, large in size and heavy in weight. With the development of miniaturization of LED optical encapsulation system, the application of secondary optical system will become more difficult. Combined with software simulation and experimental verification, the optical field regulation function of primary optical lens for the monolithically integrated light-emitting diode (MI-LED) was explored. The research results show that the optical field of experimental LED light sources is almost consistent with that of simulated LED light sources, and the designed primary lens can adjust the beam angle of LED light source from 120° to the range of 48°～72°. Compared with the LED light source without primary lens, the LED light source with primary lens has higher light extraction efficiency (LEE) and more uniform spatial light color distribution.

The multispectral imaging is an effective tool for noninvasive observation of biological tissues, and it can further accurately obtain the multispectral data of pathological tissues by improving the irradiance uniformity of the light source. Therefore, the multispectral uniform light source system based on the integrating sphere was proposed to provide uniform irradiance for multispectral data acquisition of biological tissues. First, in order to determine the optimal parameters of the structural model, the design scheme of the integrating sphere was selected based on the Monte Carlo ray tracing method (MCRT), and the cavity size, LED layout, and the output slot size of the integrating sphere were optimized. Then, an example with 11 integrated narrow-band LEDs was constructed for optical performance verification based on the optimized parameters, and the experimental results showed that all the used bands could achieve 2% irradiance nonuniformity distribution in the major part of the output slot. And the broad-spectrum compatibility of the system was studied and the experimental results showed that the maximum root-mean-square error of the spectrum in the output slot was 3.74%, which had good broad-spectrum compatibility. Finally, based on the proposed system, venous visualization tests were performed and the endoscope compatibility of the system was extended with optical fiber and used to investigate the internal dental surface pathology of the oral cavity. The multifunctionality and uniformity of the system has potential applications in biomedical image acquisition.

Interferometry is widely used in nano-scale micro-topography measurement. In order to improve its accuracy and sensitivity, a high-sensitivity homodyne interferometry based on white light interference and laser secondary interference was proposed. A high-sensitivity homodyne interferometry system was designed, and the zero point of the laser secondary interference was used to locate the dark striation of white light interference, so that it could reach the maximum slope when optical path difference was zero. The signals of white light and laser were analyzed by using the wave principle and intensity formula of interference fringes, and a sensitivity calculation method based on the combination of white light and laser interference signal was proposed. The system and its sensitivity were simulated. Finally, the optical path was built, and the white light interference fringes were adjusted to the dark striations position, so as to locate the zero position of laser secondary interference and carry out the data acquisition. It is showed that the sensitivity of the measurement method is at least 1 832 times higher than that of the laser secondary interference, and the corresponding measurement uncertainty is only ±0.288 7 mV. The measurement system can effectively solve the problem of large amount of calculation in traditional interferometry, and has high sensitivity, stability and reliability.

For monitoring the sloshing of the rocket body for liquid launch vehicle on long-term duty, and ensuring the safety and stability of the product, an on-line monitoring method of rocket body sloshing based on binocular vision measurement was proposed. Firstly, the internal and external parameters of the camera were calibrated with high precision by total station and other devices. Secondly, the precise coordinates of the same marked point at different time were measured. Finally, a set of equations were constructed to solve the parameters of pose transformation through the coordinate changes of the marked points at different times and the constraint conditions, and then the amount of rocket body sloshing was obtained. The results show that the maximum measurement error and total deviation of single marked point are 0.026 mm and 0.079 mm respectively, which can effectively meet the actual needs of on-line monitoring of rocket body sloshing.

The defects in photovoltaic cells affect the service life and power generation efficiency of the entire photovoltaic system. Aiming at the high missed detection rate of weak and small defects in the automatic detection of existing cells, a feature-enhanced lightweight convolutional neural network model was established. The feature enhancement extraction module was designed specifically to improve the extraction ability of weak boundaries. In addition, according to the principle of multi-scale recognition, a small target prediction layer was added to realize multi-scale feature prediction. In the experimental test, the mean average precision (mAP) of the model reaches to 87.55%, which is 6.78 percentage points higher than the traditional model. Moreover, the detection speed reaches to 40 fps, which meets the accuracy and real-time detection requirements.

An attitude measurement method based on multi-sensor combination was proposed to solve the problem of target feature points occlusion due to spatial obstruction in large-scale spatial attitude measurement. The single-coordinate reference measurement of the measured feature point was realized by the digital level and the attitude probe, and the initial attitude of the target was obtained by the level difference of the feature point and its known geometric constraint relationship. On this basis, the attitude rotation matrix between high-precision tilt sensor and measured target was calibrated, and then the target attitude was calculated in real time from the sensor output based on the coordinate transformation theory. The experimental results show that the relative accuracy of attitude measurement is better than 0.001 5° in the range of 10 m, and the repeatability measurement error is less than 0.000 4°, which can be used for the precise real-time measurement of large-scale obstructed spatial attitude.

The crack is a common disease on bridges and roads. Aiming at the problem that its detection accuracy needs to be improved, a bridge cracks detection algorithm based on Mask region-based convolutional neural networks (RCNN) was proposed, and a semantic enhancement module (SEM) was designed. Combined this module with feature pyramid network (FPN), a new multi-scale feature map was obtained by feature fusion. In view of the complexity and diversity of crack forms and the difficulty of identification, the cracks were divided into two categories for detection, and two strategies were formulated for comparative experiments. The results show that the improved method can get better detection results, the detection accuracy can reach to 99.8%, and the mean average precision (mAP) can be improved by 12.6%.

In order to study the influence of adjacent spectral lines under positive pressure in the gas detection process of tunable diode laser absorption spectroscopy (TDLAS) technology, a pressure measurement model based on integral absorbance was established. Taking CO2 as the research object, the simulations and experiments of pressure measurement in the range of (1~2) atm were carried out at room temperature and high purity. The results show that as the pressure increases, the degree of mutual influence between adjacent absorption spectral lines increases, and the degree of deviation of the absorbance curve from the zero baseline increases. The pressure measurement results have a maximum relative deviation of 4.94% at 1.25 atm, a minimum relative deviation of 0.73% at 2.00 atm, and an average relative deviation of 3%.