This PDF file contains the front matter associated with SPIE Proceedings Volume 12963, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

Currently, advanced aero engines are developing towards high rotational speed, efficiency, thrust weight ration, reliability and turbine inlet temperature, which needs measuring and inspection technology of high level to support the design and manufacturing procedure. Along with the development of measurement technology and other related disciplines, as a kind of non-contact measuring method, optical measuring technology could be used to cope with so many problems which may be difficult or un-solvable for those traditional inspecting manners. Further, as a branch of the optical measuring technology, machine vision uses industrial cameras etc. to distinguish, measure and locate the target and 2D of the target to derive information, which possesses such advantages as compact structure, rich information and flexible application etc. Under this background, fruitful research has been carried out in the aspect of machine vision measuring and inspecting technology of key geometrical features of aero engine such as film cooling hole, blade-shaped hole and blade defect. Also, related special equipments have been developed to realize the efficient inspection of those key geometrical features.

Laser fuze usually adopt an active detection method, which emits laser pulses to the target through the transmitting system. After scattering by the target, the receiving system obtains the return laser signal. By calculating the time delay between the transmitting pulse and the receiving pulse, the target distance can be calculated. Traditional circular-viewing laser fuze typically use four zones for 360 degree detection, with a detection field angle accuracy of only 90 degrees. In order to achieve accurate measurement of target azimuth angle by laser fuze, this paper proposes a method based on the position of laser spot on a photodetector to calculate the target azimuth angle. The effects of proposed method are analyzed by simulation experiment. The results demonstrate that this method can accurately detect the target azimuth angle.

Optical axis parallelism is critical to high-quality clear imaging and stable field switching during zooming. In this paper, we propose a fast measurement method of optical axis parallelism based on collimator and image processing technology. Our method improves the measurement accuracy by introducing skeleton thinning algorithm. The method realizes the automatic measurement of the optical axis parallelism during zooming and reduces subjective errors. The experimental results show that the measurement efficiency and accuracy have been improved, confirming the practical value of the method.

In the distributed optoelectronic system, when the optoelectronic reconnaissance equipments cooperates to locate the target, it needs to use the azimuth and elevation information of the optoelectronic reconnaissance equipments, which usually has measurement errors. This paper proposes a distributed optoelectronic system collaborative positioning model with measurement errors, and analyzes the influence of the measurement errors of azimuth and elevation on the target positioning error, At the same time, the influence of the baseline between the optoelectronic reconnaissance equipment on the target positioning error is analyzed. the influence of the included angle between and the positioning lines on the target positioning error is analyzed. The modeling analysis shows that the smaller the measurement error of azimuth and elevation angle are, the smaller the target positioning error is; The longer the baseline is, the smaller the target positioning error is; The closer the included angle of the positioning line is to 4/π, the smaller the target positioning error is. It provides a basis for the selection of angle measuring sensors in the distributed optoelectronic system and the layout of optoelectronic reconnaissance equipment in the distributed optoelectronic system.

In order to improve the anti-interference and tracking performance of photoelectric stabilized platform, a sliding mode controller based on reaching law was designed. Since the differential signal of the input was used in the sliding mode controller, if the noise was added in the input signal, its differential signal will amplify the noise, thus affecting the actual effect of the controller. To solve this problem, a method combining nonlinear Tracking Differentiator (TD) with sliding mode controller was proposed ,then the experimental system was built, and the results of the new controller and the traditional PID controller were compared. It is proved that the method this article presented can improve the anti-interference performance of the system by 66.7%, and also can increase the track precision of the input signal by 48.2%.

The imaging performance of the infrared detector is related to the key indicators of the infrared remote sensor, such as the dynamic range and accuracy of temperature measurement, and directly determines the imaging quality of the infrared remote. In order to obtain the imaging characteristics of the infrared detector in the actual remote sensors, and further improve the design scheme and production process of the infrared detector. It is proposed that based on TDI infrared detector, an imaging test verification system is built in this paper. Test results show that the imaging characteristics of the infrared detector are obtained through a series of different target image data, and the image data is quantitatively studied and analyzed. The response of each channel of the detector is not independent and affects each other which will lead to the degradation of temperature sensitivity and temperature measurement accuracy of the imaging system. For image applications it will affect the discovery, recognition and resolution of the target and accuracy of temperature inversion. The research methods of imaging characteristics based on TDI infrared detector has reference significance for the quantitative evaluation of other infrared detectors and imaging systems.

The liquid refractive index is one of the important optical parameters reflecting the properties of solution. A simple scheme based on shading effect for measuring liquid refractive index is presented. When a light beam passes through the liquid, the total reflection occurs at the interface with the air to produce a shading effect, forming a circular shading pattern, and its radius is related to the refractive index of the liquid. Based on this principle, a liquid refractive index measurement system based on smartphone is built. A rectangle colorimetric dish with tested liquid is illuminated by He-Ne laser. A white stickable label paper is attached to one side of colorimetric dish. Then the circular shading pattern displayed on label paper can be recorded by smartphone. A smartphone data processing program was developed to calculate the liquid refractive index from the image taken by smartphone. System calibration and smartphone shooting parameters are discussed in detail. The refractive index measurement experiments of sucrose solution and sodium chloride solution with different concentrations demonstrate the feasibility and effectiveness of the proposed method.

We present a simple and stable liquid refractive index measurement technique via electrically tunable lens (ETL). A focal point is produced when the parallel light passes through a convex lens. If a parallel plate with tested liquid is placed behind the lens, the position of focal point will be changed, which reflect the change of liquid refractive index. The axial displacement of focus point can be obtained by moving the camera and used to calculate the liquid refractive index. To avoid the mechanical movement of system, the convex lens is replaced by an electrically tunable lens. By exactly controlling the external currents of the ETL, the position of focal point can remain unchanged after placing the parallel plate with tested liquid. The change of focal length can be obtained by the change of external currents, which equal to the axial displacement of focus point. Thus, the liquid refractive index can be measured with the change of focal length. Some experiments of water, blend oil and 75% alcohol for measuring the liquid refractive index demonstrate the feasibility and effectiveness of the proposed method.

In this paper, a monolithic polarizing metalens is used to extract multiple polarization information of the target at the same time. The i1maging receiver uses a general detector easily available on the market, which can obtain the intensity information image of multiple polarization components. The polarization measurement speed is fast, and the number of modules in the optical system is reduced The monocular 3D imaging system based on polarization-sensitive metalens collects diffuse reflected light from the surface of the target, and uses different polarimetric sub-images obtained by the metalens to carry out 3D reconstruction of the target.

Diffractive imaging techniques, such as coherent diffraction imaging (CDI), ptychography, and Fourier ptychography, have gained popularity due to their ability to recover the amplitude and phase information of samples simultaneously from the diffracted pattern with super resolution and wide field of view. However, imaging noise can significantly degrade the reconstructions in diffractive imaging. Higher order diffractions, in particular, are sensitive to measurement noise due to their lower signal-to-noise ratios (SNR) compared to lower orders. Existing denoising methods cannot effectively separate signals from detector noise. To address this limitation, we propose a self-adaptive noise minimization approach using a regularized regression method. Our approach involves training a regularized linear regression model to evaluate the power of noise level in the recorded noisy diffraction patterns and the detector's dark noise. This results in a refined pattern with high SNR. We evaluate our approach on synthetic and experimental datasets and compare it with existing noise reduction methods. The results demonstrate that our method significantly outperforms other state-of-the-art methods in terms of both noise reduction and preservation of fine structural details. Moreover, our approach does not require any prior knowledge or assumptions about the noise statistics, making it a robust and versatile method for diffractive imaging applications.

An optical fiber-type fluorescence sensor based on surface-enhanced fluorescence (SEF) is proposed and analyzed through the finite element method. The field enhancement property of the designed multimode SPR fiber is utilized to improve the excitation rate of fluorescent molecules, which improves the sensitivity of fluorescence detection. Because the gold film is too close to fluorescent biomolecules, which can lead to fluorescence quenching, an optically transparent layer is added between the surface of the gold nanofilm and fluorescent biomolecules. And then analyze the layer's material and thickness parameters. When the gold film and PDMS thicknesses are 50 nm and 60 nm, the field enhancement factor (T) reaches the maximum value of 9.45. This fluorescence enhancement method provides a design possibility for improving the fluorescence detection sensitivity.

Optical remote sensing satellites are an important part of the current Earth observation technology, and with the development of space technology and the diversification of application requirements, their imaging models are becoming more and more diversified. The overall development trend of optical remote sensing satellites is to achieve ultra-large width and ultra-high spatial resolution. In the face of the growing demand for large-scale and high-timed Earth observation in the region, rotational geography has gradually become a new model and has been gradually studied and developed. The rotating ultra-large wide imaging optical satellite adopts the conical scanning earth observation technology system of area scan camera, realizes ultra-large area observation through conic scanning, and designs widearea search imaging and encryption imaging modes in key areas, which has the characteristics of wide-area, highresolution and multi-angle imaging. The degradation information of rotational imaging optical satellite images is difficult to accurately estimate and the spatiotemporal characteristics are complex, and the ground imaging process is composed of satellite flight motion and turntable rotation, and in order to compensate for the image shift effect in the imaging process, the satellite is equipped with a fast reflection compensation mirror and the detector rotation mechanism for realtime motion compensation, which makes it difficult to accurately construct and calculate the image degradation model. In order to accurately obtain the mechanism and process information of image quality degradation at the focal plane, the accurate perception and modeling method of dynamic imaging degradation information is studied, and key technologies such as progressive information fusion based on overall weighting of bidirectional filtering, separation and spatiotemporal modeling of high and low frequency degradation information based on Fourier model, and single-sided image compensation restoration based on degenerate spatiotemporal model are broken, and distortion-free high-quality images in rotational maneuvering mode are obtained, which lays a theoretical and algorithmic foundation for the design and data processing of rotating imaging optical satellites.

Currently, China is still a major consumer of coal resources. Coal can be used in various fields such as industry and civil use, and can be used for power generation, heating, and building materials. There are many types of coal, each with its unique composition and properties. It has specific requirements for its use in various fields, which make the use of coal more reasonable and important for the sustainable development of the environment and resources. Therefore, the classification research of coal is of great significance. Due to the same component influence among various coals, there are certain challenges for coal classification. Therefore, a laser induced breakdown spectroscopy (LIBS) based on principal component analysis (PCA) combined with convolutional neural network (CNN) method was proposed to classify and recognize coal samples from six different regions. Through laser ablation of coal samples and collection of corresponding data, the data are dimensionalized and standardized, and then the spectral data are classified and trained through PCA-CNN optimization model. The final results indicate that the coal classification accuracy of the PCA-CNN deep learning network model can reach 98.15%. From this result class, it can be seen that laser induced breakdown spectroscopy technology combined with PCA-CNN can achieve rapid and accurate classification of coal samples from different regions, and provide a new coal quality detection data analysis and processing scheme.

The interfacial asymmetry and compositional disorder caused by atomic segregation and exchange significantly affect the electrical properties of the InAs/InAsSb superlattice, leading to deviations from original designs. The study presents a quantitative analysis of the compositional asymmetry of the superlattice and its effects using a segregation model and 8-band k.p model. The composition disorder at each interface, primarily induced by Sb segregation, is examined through the reconstruction of the actual atomic sequence structure based on scanning tunneling microscopy results. Three different atomistic structures of the superlattice are modeled using the k.p method, including the ideal MBE-growth structure, a rebuilt structure with Sb segregation only at the InAs-on-InAsSb interface, and a rebuilt structure with Sb segregation at both interfaces. The results of the modeling highlight the significant influence of Sb segregation on the electronic properties of InAs/InAsSb superlattices.

Ultrasound has strong penetrability in opaque media, which makes it one of the most important detection tools for structural safety monitoring. Compared with the traditional piezoelectric transducers(PZTs) used for ultrasonic detection, fiber-optic ultrasonic sensing system based on optical adaptivity of two-wave mixing (TWM) photorefractive crystal has strong characteristics of anti-electromagnetic interference, high sensitivity and strong multiplexing, as well as the adaptive detection for optical phase modulation and permanent dynamic recording of the interference pattern current position, which has a large quantity of applications in structural health monitoring. For the interferometric properties and ultrasonic detection technology of TWM in the field of structural safety monitoring, this paper introduces the principle of adaptive two-wave mixing interferometry, the sensing devices, and methods in various fiber Bragg grating (FBG) sensor systems and dynamic signal detection of Er-doped fiber gratings and discusses the application of fiber-optic ultrasonic sensing systems in ultrasonic detection based on adaptive TWM technology. Finally, the problems and solutions in the sensing system are analyzed. By miniaturizing the TWM interferometer and integrating the entire sensing system, it is expected to further improve the detection performance of the device and expand future research directions.

In recent years, there has been a tremendous development in the computing and processing power of mobile platforms, leading to a continuous emergence of research and achievements in mobile platform image processing. The processing of microscopic images using mobile platforms has also become an increasingly urgent demand. One important goal of image processing is to enhance the quality of the images, and in the case of microscopic images, color images make up a significant portion. Therefore, it is necessary to implement a microscopic image enhancement algorithm on mobile platforms to further improve the observation and capturing experience of microscopic images on these devices. In this study, extensive research work was conducted to analyze and study the implementation of image dehazing algorithms. By considering the characteristics of microscopic images, the algorithm was improved to achieve a mobile platform-compatible microscopic image enhancement algorithm, which was successfully applied on an Android mobile platform with Rockchip's RK3399 as the core. Extensive experiments were conducted on slice samples under different magnification levels and lighting conditions, demonstrating that the algorithm can achieve efficient and outstanding microscopic image enhancement.

Filament refers to a plasma channel with high laser intensity and high plasma density formed by propagation of intense femtosecond laser pulses in transparent medium. Direct experimental measurement of a filament is often impossible due to the extremely high speed of the process (~ps) and sizes of studied objects (~μm). In this paper, a novel medical imaging method named photoacoustic tomography (PAT) is used to reconstruct the cross-sectional image of multiple filamentation of femtosecond laser pulses in air. An open toolbox named k-Wave is used to simulate the acoustic signal generated by optical filament with different energy deposition density. The conventional delay-and-sum (CDAS) algorithm is used to reconstruct the initial pressure distribution of the multifilaments. This study verifies the feasibility of using photoacoustic tomography to reconstruct femtosecond multifilament images, which can provide theoretical support for the practical application of optical filament in atmosphere.

The modular segmented space telescope can get rid of the limitation of the effective envelope of the fairing of the existing launch vehicle, meet the requirement of the large aperture of the space telescope, and play an important role in space exploration and earth observation. In order to solve the problem of power supply and signal transmission of the primary mirror of the segmented space telescope, a kind of modularization space telescope primary mirror system for in-space assembly is first proposed in this paper. Then, according to the index requirements and the assembly process of primary mirror, the wireless energy and signal transmission scheme and system structure are proposed. By analyzing the key parameters, the optimal system of wireless energy and signal transmission on the segmented mirror is achieved. The proposed system realizes high power and high efficiency transmission of power supply energy and reliable transmission of data signal during and after assembly of the segmented mirror, and solves wire connection problems caused by the traditional wired transmission mode.

This paper first introduces the application background of detection and processing integrated intelligent image sensor, and proposes an overall architecture of intelligent image sensor that uses hybrid stacking process for vertical interconnection. Among them, the top pixel layer mainly contains 10K×10K 5um pixel array and row and column drive array, which gives the pixel structure, simulates and analyzes the pixel noise index and effectively guides the design of the readout circuit structure. The lower chip layer mainly contains readout circuits, image signal processing ISPs, general-purpose CPU units, AI neural processing units, and on-chip SRAM. The image signal processing ISP function and the AI neural processing unit are introduced in detail, and the FPGA verification results of the AI neural processing unit are given.

Aiming at the problem of reconnaissance and positioning of ground targets by airborne photoelectric detection equipment, this paper firstly studies the problem and model of multi-cooperative positioning based on distance measurement, and analyzes the factors affecting the positioning accuracy. Then, three solving methods of multicooperative target location based on distance measurement are proposed: least square algorithm, traditional particle swarm optimization algorithm and improved particle swarm optimization algorithm based on least square. The positioning accuracy and efficiency of the above three algorithms are simulated and compared in MATLAB, and the above algorithms are verified by flight test data. The experimental results show that the improved particle swarm optimization algorithm based on least squares has high computational accuracy and efficiency for solving the location equation of cooperative distance measurement.

Influenced by the prominent progress of two-dimensional (2D) layered crystals, the fabrication of 2D nanostructures from non-layered materials has attracted more and more attention. Lead selenide (PbSe) is one of the superior candidate materials for photodetector with suitable bandgap and outstanding photoelectric properties. The growth and device preparation of PbSe supply great interest for the development of high-performance infrared photodetectors. Although a lot of efforts have been paid on preparing PbSe nanostructures for miniaturized detectors, it is challenging to synthesize excellent crystallinity and thin 2D PbSe nanosheets because of itsinherent rock salt nonlayered structure. In this work, we employ a catalyst-free facile physical vapor deposition (PVD) method for controllable synthesis of PbSe nanosheets by van der Waals epitaxy technology. By optimizing the growth temperature, PbSe nanosheets from triangular pyramid island to square 2D plane can be obtained. In addition, the 2D PbSe nanosheets detector has a responsivity of 3.03 A/W at the wavelength of 520 nm with the power density of 5.05 mW/cm². This work provides a facile strategy to synthesize high-quality 2D PbSe nanosheets which have enormous potentials to fabricate high-performance miniaturized photodetector.

Theoretical analysis and experimental research are carried out on the signal processing of pyroelectric sensor D203S used in infrared carbon and sulfur analyzer. The circuit system formed, through simulation and actual measurement, optimizes the circuit parameters, and obtains experimental results that meet the requirements of bandwidth and signal-to-noise ratio.

Due to the increasing demand for miniaturization and portability, the development of self-powered photodetectors that can work without external power supply has aroused great interest among researchers. As a group-10 layered transitional metal dichalcogenides, PtSe2 exhibits potential applications in photoelectric detection because of the unique properties such as high carrier mobility, tunable bandgap, and stability. However, its inherent large dark current hinders the further improvement of the performance of the PtSe₂ photodetectors. In this paper, we fabricated a vertically aligned Two-dimensional (2D) van der Waals (vdWs) heterojunction composed of PtSe₂ and MoSe₂, which exhibits high sensitivity photoelectric detection performance in a wide band from visible light (405 nm) to near-infrared (1550 nm) without external bias. As a result, working in self-driving mode at room temperature, the responsivity and detectivity can reach 22.95 A W^-1 and 9.27×10¹¹ Jones with a fast response speed of 180/48 μs. This work is expected to provide a new idea for broadband, energy-efficient and high-performance miniaturized detectors.

It needs a triggered time to open or close optical gate of the ICCD. The duration time is defined as irising effect time. Although it only lasts a few nanoseconds and often be ignored by users, it can still interfere with the results in some applications such as fluorescence lifetime imaging and gated imaging. This paper proposes a fitting algorithm to correct the irising effect of the image intensifier. This method obtains fitting matrices of different gates through a series delay images of ICCD. Then through these fitting matrices, the imaging pictures are effectively corrected. The advantages of this method are low cost, high efficiency, and simplicity. The verification experiment of this paper is to write letters with a highlighter. The correction algorithm can clearly restore time-resolve image, and significantly improve the contrast of the image.

The two key factors in super-resolution image reconstruction are the preprocessing of LR image sequences and the selection of reconstruction algorithms. Due to a series of factors such as noise and blur that may affect the captured image sequence, it is necessary to preprocess the image sequence first. For frames that cannot be processed, they should be considered as bad frames and removed to prepare for reconstruction. The selection of reconstruction algorithms is crucial, and the standard for a good reconstruction algorithm is fast reconstruction speed and good reconstruction results. This article compares the advantages and disadvantages of reconstruction algorithms both domestically and internationally, It is found that the POCS method is more suitable for this article, so more in-depth research and improvement have been conducted on the POCS method. The experimental results show that the improved POCS reconstruction algorithm has a relatively fast reconstruction speed and good reconstruction effect.

The ferroelectric liquid crystal (FLC)-based wide spectrum polarization imager can quickly acquire the full polarization information of the target, however, the polarization elements error will affect the accuracy of the system polarization parameters (DOP, EOP, and AOP) measurement. In this paper, to investigate the influence of polarization element error on the acquisition of polarization information of the measured target, the influence of element parameter error on the measurement of target polarization parameters is analyzed. First, the working principle of FLC and the structural composition of the system of wide spectrum polarization imager based on FLC are introduced; second, the measurement matrix of the wide spectrum polarization imaging system is calculated based on the CCD spectral response curve, and the variation of the wave plate phase retardation with wavelength; then, the error model of the system for solving the full polarization parameters is established; finally, the effects of the parameter (azimuth angle F1θ, tuned rotation angle F1α, phase retardation F1δ of FLC2; azimuth angle Hθ, phase retardation Hδ of HWP; azimuth angle F2θ, tuned rotation angle F2α, phase retardation F2δ of FLC2; azimuth angle Qθ, phase retardation Qδ of QWP; azimuth angle Pθ of P) errors of each element of the polarization module on the DOP, EOP, and AOP measurements are analyzed. The simulation results show that: with the increase of the error of each element parameter of the polarization module, the measurement error of the polarization parameters are all increasing in different degrees; the error of F2α is the main factor affecting the measurement accuracy of the degree of polarization (DOP), the error of F1α is the main factor affecting the measurement accuracy of the ellipticity of polarization (EOP), and the error of F1θ is the main factor affecting the measurement accuracy of the angle of polarization(AOP); at [-2°,2°], the maximum measurement errors of DOP, EOP, and AOP are 2.83%, 6.98%, and 9.26%, respectively; when the errors of polarization element parameters are certain, the measurement errors will change with different polarization states of the target. The research method and results can provide research ideas and theoretical references for the error analysis research of the wide spectrum polarization imaging system based on FLC or other wide spectrum polarization imaging systems.

Aiming at the high-precision pointing requirements of on-orbit splicing or formation, a high-precision pointing control based on visual pixel deviation is proposed. First, the telephoto camera is used to identify the cooperative target at the sub-pixel level, and the pixel deviation of the target center in the center of the camera field of view is used as the pointing control amount, and the relationship model between the target pixel and pointing angle is established; on this basis, the momentum wheel is used as Output torque, the control model of pixel offset and output torque is established, and relatively high-precision pointing control is realized. The pointing control experiment of the single-axis air bearing turntable shows that the pointing angle resolution of a single pixel of the camera reaches 0.0037°, the pointing control accuracy reaches 0.0091°, and the pointing control stability reaches 0.01°/s, which verifies the high accuracy and effectiveness of the control scheme sex.

The server in the computer room runs in a closed, narrow, and uninhabited environment. Due to the temperature rise and heat dissipation of the device, faults, burns out, or even fires may occur due to high temperature. At present, the traditional temperature monitoring methods in computer room are limited to electrical method, infrared thermal imaging detection method, fiber Bragg grating measurement method, rarely involving high-precision and continuous measurement methods. By introducing optical fiber on the surface of the server body, the author designs and builds the ROTDR distributed optical fiber sensing system for temperature measurement in the computer room, realizing the integrated direct measurement. The system has the advantages of high precision, strong stability and fast response time, which provides a reliable theoretical basis for temperature safety monitoring and fault warning in the computer room.

With the development of remote sensing technology, users' demands for agility and high resolution are becoming increasingly high. Beijing No.3 Satellite is a new type of optical remote sensing satellite with ultra-high agility, high resolution and large swath. Operating in about 500kmSun-synchronous orbit, it can obtain the ground pixel resolution of panchromatic 0.5m/multi-spectral 2m at the satellite nadir point with a width of more than 23km. According to user indicators and engineering implementation difficulties, a dual camera optical splicing method is adopted to increase the camera width and reduce the development difficulty of the camera's optical and mechanical structure. The splicing process adopts the combination processing method to ensure the high precision and stability of the splicing angle of dual cameras. In terms of opto-mechanical structure, the stability is ensured through the integrated design of the mirror and the support structure. Thein-orbit commission shows that the camera resolution and swath meets the users’ requirement, the image is clear and the quality is high, indicating that the camera's optical and mechanical structure has high stability and demonstrate the effectiveness of the approaches proposed.

In the process of scene based non-uniformity correction, the scene is often changeable, and the corresponding detector operating range of different temperature scenes is different, and their nonuniformity parameters are also different, so when the scene changes, the non-uniformity parameters will also drift, and it is easy to introduce serious ghosts in the correction results, which greatly reduces the Rate of convergence of the algorithm, This poses serious problems for real-time non-uniformity correction algorithms. At the same time, the scene is very flexible, and a single model cannot describe the scene well. Therefore, on the basis of treating the dynamic change process of the scene as a Markov process, this article proposes an adaptive non-uniformity correction algorithm based on Multi Model Particle Filter (PF-NUC). By introducing a tracking framework of particle filter, a nonlinear and non Gaussian parameter estimation model is established. Through experimental simulation verification, after the algorithm proposed in this article, almost no ghosts or residual non-uniformity can be seen. Through visual evaluation, the PF-NUC method has the best ability to remove fixed pattern noise.

Computational optical microscopy can obtain multidimensional and microscopic sample images and information that cannot be observed by the naked eye. However, due to limitations in hardware manufacturing level, optical deployment methods, and detection tools, there are often errors between the actual imaging system and its physical model, which makes it difficult for computational optical microscopy to achieve stable imaging functions and obtain high-quality observation images. This paper introduces an automatic correction method for microscope stage based on Hough Transform image processing. It does not require manual operation and only requires image processing of the imaging results to complete the correction fully automatically. The correction process is simple and fast, with better visualization and accuracy for the imaging results. It can greatly improve the efficiency of stage correction and ensure that the computational optical microscopy system can obtain high-quality observation images, and it can be universally applicable to optical microscopy systems with electrically controlled displacement stage.

In the field of autonomous navigation, the accuracy of the inertial navigation system (INS) heavily relies on the initial position information. Typically, this information is obtained from the global navigation satellite system(GNSS) or celestial navigation system (CNS). In this study, we present a novel approach for acquiring precise initial position information using the INS/CNS integration system. This method enables the independent determination of the carrier’s high-precision initial position in a static state. The experiment results show that the latitude error is 48.35m and the longitude error is -55.80m of this approach, which provide compelling evidence for the effectiveness of the proposed celestial positioning method.

With the rapid advancement of space astronomical observation technology, there is an increasing demand for higher camera imaging quality. The micro-vibrations generated by internal disturbances of the satellite platform are the primary factors affecting encircled energy of disc of confusion in star image point. The traditional optical image stabilization system compensates for the disturbance by acquiring the variation of the centroid coordinates of the image point as an error feedback of the system, which has the problems of obvious lag and insufficient rapid response ability. To overcome the problem, a method based on optical image stabilization system for predicting angular velocity was proposed and on the basis of measuring the angular velocity of the Satellite bus as the error feedback of the systemin this paper. The measured data are predicted using a combination of current statistical model (CS) and Kalman filtering algorithm. The simulation results in MATLAB demonstrate that the image stabilization system's line of sight stabilization is better than1".It is shown that the method proposed in this paper has feasibility in the optical image stabilization system, the proposed algorithm has the characteristics of high robust and low computing load, and can accurately and effectively predict the vibration information of satellite platform in advance, and provide help for the rapid implementation of vibration suppression measures.

A non-uniformity correction method based on machine learning for A/D driver circuit level. Firstly, different types of infrared detectors are placed in a temperature uniform radiation field. When they are working normally, the analog output signal waveform of each detector in multiple scenarios is collected multiple times to obtain the approximate average voltage value of each line in a frame, and saved as a document; Secondly, using GPU for machine learning of the above documents and accurately driving the D/A chip for digital to analog conversion, simulating the waveform voltage value of the analog output signal mentioned above to generate waveform voltage with similar nonuniformity; Thirdly, the voltage waveform generated by the multi-channel voltage output digital to analog converter is followed by an operational amplifier, filtered, and then output to the A/D chip as the reference voltage for sampling; Finally, the analog video signal output by the infrared detector is sampled and quantized by an A/D chip to obtain a more uniform image digital signal.

The interworking between different networks can be effectively realized by implementing all-optical format conversion, and it can improve the flexibility, efficiency and expansibility of the communication system. The all-optical signal processing scheme is only carried out in the optical domain, and it does not require optical-electric-optical (O-E-O) conversion, which greatly improves the transmission rate and processing speed of the communication system. In this paper, an all-optical format conversion scheme from dual-polarization on-off keying (DP-OOK) to quadrature phase shift keying (QPSK) is proposed. The scheme is based on vector phase-sensitive amplification (PSA). Constellation and eye diagram of signals at each stage of system are obtained by simulation, and BER and EVM are calculated to show the performance of the system.

Underwater three-dimensional gated range-intensity correlation imaging (3D GRICI) can obtain two-dimensional images with large target-to-background contrast by suppressing backscatter and background noise outside volume of interest (VOI), and simultaneously reconstruct 3D images by the range-intensity correlation algorithm. However, it is still affected by the sub-backscatter noise from VOI in turbid water, resulting in short 3D detection distance and low contrast images. Therefore, in this paper the optical polarization is used for 3D GRICI. Due to the polarization-preserving property of water backscattering, the sub-backscatter noise can be removed from gate images by polarization. Experimental results show that when the water attenuation coefficient is less than or equal to the critical attenuation coefficient of c0, the polarization has no effect on improving detection distance and imaging quality of 3D GRICI; when the water attenuation coefficient is greater than c0, the polarization is helpful for improving performance of 3D GRICI. This research is conducive to optimize the applications of 3D GRICI in turbid water.

In order to improve the real-time performance of the division-of-time polarization imaging system applied in dynamic scenes, the traditional rotating polarizer imaging scheme is improved and a fast-rotating polarization imaging system is designed. The scheme of the camera exposing during the uniform rotating of the polarizer is applied in the system, and the intensity images of different polarization angles are collected when a designed hollow turntable drive the polarizer to rotate rapidly. Based on the brief introduction of the principle of the polarization imaging system of fast rotating polarizer, the optical-mechanical structure design and core components of the system are introduced in detail, and the polarization imaging experiment of the system is carried out. The pipeline calculation method and the least square method are used to solve 5 adjacent intensity images of the target scene every time to calculate the degree of linear polarization (DoLP) and the angle of polarization (AoP). The test results show that the system can get polarization images at a frame rate of 80 frames per second (FPS) or with the pixels of 1280×1180. It is shown that when the imaging frame rate is 51 FPS, the polarization images obtained by solving 5 consecutive frames of intensity images have better detail recognition ability than those obtained by solving 3 consecutive frames of intensity images. Compared with the division-of-time polarization imaging system of the start-stop rotating polarizer, this division-of-time polarization imaging system with fast rotating polarizer increases the polarization imaging frame rate, improves the real-time performance, and enhances the detection ability to dynamic scenes.

In the field of computer vision, point target tracking has always been an important topic and research hotspot, and it is widely used in both military and civilian fields. For the tracking of point targets under complex background, the point targets are extremely small, and their morphological characteristics are not obvious, so they are easily disturbed by background and noise. Secondly, the point targets’ maneuvering, shaking of detection equipment, etc., will change their morphology, resulting in low detection rate and high false alarm rate, which will further affect the accuracy and robustness of point target tracking. Therefore, how to effectively utilize the spatio-temporal information in sequence images to extract the target accurately is a difficult problem. This paper summarizes the existing detection and data association algorithms in point target tracking, analyzes their performance and shortcomings, and discusses the development direction of point target tracking algorithm, that is, algorithms based on multi-feature fusion with strong robustness, high accuracy and small calculation.

In the field of face anti-counterfeiting, there are differences between the reflection spectrum of real faces and simulated faces, which can help us overcome the shortcomings of traditional RGB cameras that are difficult to identify the authenticity of faces. In our work, we designed a face anti-spoofing imaging system based on the snapshot spectral imaging chip, which can be used in face anti-spoofing imaging through the analysis of spectral imaging data. Experiments show that our sensor could reconstruct the spectrum of the face region, establish the spectral databases, and achieve face authenticity recognition under active light source based on deep convolutional neural network, with a face recognition accuracy of 95%.

To meet the demand of weak ultraviolet signal acquisition and transmission for solar-blind ultraviolet flame detectors, this paper develops a signal acquisition system based on a β-Ga₂O₃ thin-film photoconductive sensor. Firstly, the weak signal acquisition circuit is designed according to the electrical characteristics of the β-Ga₂O₃ photoconductive sensor. Secondly, real-time data transmission and processing are achieved by using field-programmable gate array (FPGA). Thirdly, the experimental platform of solar-blind UV imaging system is built, and the joint test of the β-Ga₂O₃ photoconductive sensor and the readout electronics system is conducted. The experimental results demonstrate that the system is capable of acquiring, transmitting, and storing the solar-blind UV signal data from a 1×8 pixel array. The sampling rate can reach up to 20 kHz and exhibits stable performance.

With the continuous development of the field of remote sensing, the field of space-based remote sensing is developing in the direction of all-sky time and intelligence. Since low-light remote sensing is used to detect ground objects under low illumination conditions such as night and dawn and dusk, it leads to the characteristics of low contrast, low brightness and low signal-to-noise ratio of remote sensing images. The low signal-to-noise ratio leads to a large number of complex physical noises which will drown the image features and seriously affect the recognition and interpretation of ground objects. In this paper, a real physical model construction method based on physical mechanism of low-light image is proposed, By solving the noise parameters of space-based low-light remote sensing camera, the specific noise distribution model is constructed, which is used to synthesize the training data of the training denoising network, Thus, it gets rid of the difficult and laborious calibration problem caused by the lack of space-based data. In addition, in order to simulate the real imaging scene of space-based remote sensing camera, a set of low-light remote sensing noise image data set based on real physical model is constructed for the first time, which effectively promotes the development of subsequent image denoising research.

Complicated environment and obstacles like power lines are pernicious to low-flying helicopters. A novel obstacle detection algorithm for onboard LiDAR evadible system is proposed, which lowers false alarm rate (FAR) and raises detection speed (DS). Firstly, the pass-through filter and planarization voxel filter are used for denoising and sampling reduction. Then, the point cloud is separated by adaptive threshold elevation filtering. Finally, the power lines are extracted with line feature constraint. Experimental results show the maximum detection distance for power lines is up to 800 m, recall rate, over 95%, false alarm rate, below 5%, and that the detection time is less than 100 ms.

Intelligent optical sensing technologies play important roles in many fields, one of which is to help unmanned devices such as UAVs, autonomous mobile robots and intelligent robots to achieve accurate localization and mapping. With the advancement of Industry 4.0 and intelligent manufacturing, the use of autonomous mobile robots has become an important indicator of a country's industrial modernization. As the core issue in the research of autonomous mobile robots technology, autonomous localization and mapping technology has been the focus and difficulty of many scholars at present. Through the efforts of early researchers and engineers, the localization and mapping technology of autonomous mobile robots in simple static environment has achieved fruitful results, and is also playing an important role in the practical industrial application of autonomous mobile robots. However, when the autonomous mobile robots are faced with more complex or changing surrounding environment, the traditional localization and mapping methods based on geometric features such as points and lines can not achieve more accurate results, and even produce many wrong data to hinder the normal operation of the autonomous mobile robots. In this paper, combined with the characteristics of the complex dynamic environment that autonomous mobile robots will encounter in actual work, we propose a method to obtain and utilize the relatively advanced semantic information in the surrounding environment and use it for autonomous mobile robot localization and mapping. The method of this paper uses deep learning technology to mine more advanced semantic information based on the traditional method obtaining the geometric information of the environment, so that the autonomous mobile robot can generate advanced recognition and cognition of the objects inthesurrounding environment, thus assisting it to complete more accurate localization and mapping.

To overcome the limitations of conventional long-focus camera calibration, this paper proposes an improved algorithm for calibrating multi-perspective long-focus cameras based on the parallel perspective projection model. The algorithm employs an affine approximation projection model to describe the imaging process of long-focus cameras. By capturing images of the same static scene from different viewpoints and utilizing multi-view images and bundle adjustment, the algorithm jointly estimates camera parameters, including focal length, distortion coefficients, and rotation matrices, facilitating rapid, flexible, and accurate calibration of long-focus cameras. Finally, the proposed algorithm is compared with the Zhang Zhengyou calibration algorithm by calibrating a Canon EOS 60D18-200 camera with a 100mm focal length. The precision and stability of the proposed algorithm are verified through comparison, laying a foundation for subsequent tasks in computer vision, such as 3D reconstruction and image fusion.

Large aperture space telescope is a significant scientific instrument to achieve high resolution remote sensing and high sensitivity detection. The current approach to space telescope is designed to meet the volume and mass requirements associated with a single launch vehicle, one potential approach that addresses these current challenges and achieves that future is In-Space Assembly. First of all, the development status and technologies of In-Space Assembly Telescope (ISAT) are reviewed, and some typical projects are presented to show the future application. Then, the key technologies involved are analyzed from multiple aspects: large truss structure, robot technologies and lightweight segmented mirrors. On this basis, several potential technical difficulties and challenges in the future are given, which need to be further developed to promote the capability of space technology. Finally, the future development trends and technologies of ISAT are prospected, which may provide a reference for evolution of large aperture space telescope.

Steel structure has been widely used in modern buildings due to their excellent building performance, but their poor fire resistance requires special coating which can resistant fire for protection. With the increasing demand for fire-resistant coatings for steel structure, fire-resistant coatings with different fire-retardant mechanisms have emerged one after another. The market complexity has increased, and there is a significant difference in fire resistance performance among products, requiring real-time and effective supervision by the relevant department. However, due to the wide variety of fire-resistant coatings and the complexity of standard detecting methods, it brings difficulties for fire supervision and on-site inspection of fire-resistant coatings. The research proposed an efficient method based spectral analysis to detect the fire-resistant coatings with different fire resistive mechanisms, including both intumescent coatings and non-intumescent coating. Principal component analysis is used to quickly identify the spectral consistency of different coatings. The experiment selected samples of fire-resistant coatings with excellent performance verified by standard detection methods and ordinary one for visible to shortwave infrared 400-2500nm spectral collection and spectral feature analysis. The experimental result indicates that samples of high-performance fireproof coatings have high consistency in spectral feature. Through intelligent recognition algorithms, coating samples with unsatisfactory performance can be quickly and accurately detected. The research has shown that the intelligent spectral imaging technology is expected to provide a reliable basis for rapid on-site identification of fire resistive coatings for steel structure.

The optical imaging technology possesses characteristics such as intuition, high resolution, and convenient application, which are of significant importance in the fields of physics, materials science, biology, and medicine. Traditional imaging systems are large in volume, heavy in weight, and structurally complex. Recently, the development of Multi-level Diffractive Lens (MDL) has introduced advantages such as small volume, light weight, and easy integration. However, the current analysis of MDL imaging quality has generally been limited to a narrow spectral range, and the aperture sizes of the components are mostly in the millimeter range, which somewhat restricts their applications. This study focuses on the research of the imaging performance of a 5cm aperture MDL. To analyze the focusing performance and imaging quality, we designed and constructed an optical setup that is compatible with a 10cm aperture. The tested wavelength range covers 0.4 to 1.1 μm. The test results show that the RMS value of the focal length for the large aperture MDL is 22.67cm in the visible and near-infrared wavelength ranges. Compared to the traditional Fresnel lens, achromatism performance is significantly improved. The RMS value of the focal spot size in the visible wavelength range is 24.01μm, while in the near-infrared wavelength range is 26.64μm. The full-band imaging test results in the range of 0.4to 1.1μm show that the large aperture MDL has a line pair resolution of 161.01 lp/mm and a linewidth resolution of 3.1lp/mm. The aforementioned experimental results demonstrate the wide range of potential applications for the large aperture MDL in fields such as microscopic analysis, optoelectronic detection, and medical

The alignment of mask and wafer is a very important step in the process of lithography. When a specific pattern in the exposure image is selected as the alignment mark, the traditional automatic alignment methods which are based pre-set markers and image processing are not suitable. To address this issue, we propose a novel accurate image recognition method for arbitrarily selected mark patterns, which combines Support Vector Machine (SVM) with feature extraction to achieve adaptive switching of alignment templates. Firstly, based on the distinct linear contour features of silicon wafer exposure patterns, which lack color and texture characteristics, we extract Histogram of Oriented Gradients (HOG) features from the images to construct feature vectors ; then, we select the optimal SVM kernel function through experimental comparisons, and select regions of interest on silicon wafer exposure images for testing ; finally, we utilize HU-based shape features for secondary matching and recognition decisions. The experimental results demonstrate that the proposed method achieves a recognition accuracy of 100%, enabling the implementation of adaptive alignment template selection and switching.

To achieve precise positioning and accurate recognition of dot-distribution encoding points while enhancing efficiency, this study proposes an approach combining improved grayscale centroid and point pattern matching. The algorithm identifies encoded points using multiple constraints and optimizes grayscale centroid via dynamic threshold calculation. It accurately extracts encoded points centers. An advanced clustering method reduces parameter sensitivity and a novel decoding method employs geometric and computer vision principles. Experimental outcomes demonstrate sub-pixel accuracy, robustness to angles, and a 90% correct recognition rate at a 60-degree angle, holding practical significance.

Distributed optical fiber acoustic sensing system (DAS) can achieve real-time monitoring of vibration events along the optical fibers by demodulation of back Rayleigh scattered light information in optical fibers. DAS system has a long detection distance, low cost, high sensitivity and other characteristics, and has been playing an important role in various engineering applications. However, in practical engineering detection of DAS system, there are other vibration sound sources in the environment besides target vibration events. However, the number and type of environmental sound sources are usually unknown. This kind of uncertain multi-source interference will have a serious impact on the monitoring of target events.

A photo-counting imaging sealed detector has been fabricated for far-ultraviolet (FUV) radiation detection, including an MgF₂ input window, CsI photocathode, V-stack MCPs, circular arc terminated (CAT) resistive anode, and detector tube. The CsI photocathode was prepared by vacuum thermal evaporation. The CAT resistive anode was fabricated using thick-film technology. The detector body is mainly made of Kovar alloys and ceramics and was sealed in a cathode-transfer and indium-sealed vacuum system. The detector performance was evaluated using an FUV parallel light system consisting of a deuterium lamp, a reflective condenser, a filter wheel, a monochromator, a collimation system, and a vacuum chamber. The test results show that the imaging RMS nonlinearity is less than 3%, the spatial resolution is about 235um, and the dark count rate is less than 2.74 counts·s^-1.

To address the need for efficient and large-scale fabrication of multi-level diffractive lenses (MDL) for focusing and imaging applications, this study investigates the limitations of small aperture, high cost, and low efficiency in existing MDL fabrication processes. By analyzing the advantages and characteristics of various techniques in the micro-nano structure fabrication process, we propose an MDL fabrication method that combines grayscale photolithography and nanoimprint lithography. Experimental results demonstrate that the proposed method enables precise fabrication of MDL microstructures with a 10% error margin, exhibiting excellent structure control capabilities and fabrication accuracy. Notably, compared to traditional processing methods, the fabrication time for MDLs has been reduced from 5 hours to 10 minutes per piece using the method suggested in this study. By integrating nanoimprint lithography and grayscale photolithography, we successfully achieve high-efficiency and high-precision replication manufacturing of large-aperture MDLs. In conclusion, this research offers a high-efficiency and high-precision replication manufacturing method for the mass production of MDLs, potentially promoting the industrialization of MDLs in optical application fields.

In terms of public security, voice is one of the most crucial types of intelligence, and research on the remote collection of voice intelligence has substantial research ramifications and applications in fields like counterterrorism and narcotics control. Most optical remote sensing-based laser voice acquisition technologies in use today illuminate static target objects placed in a long-distance sound field using invisible infrared lasers, and then use phase demodulation to obtain vibration information of the target along the beam direction. The benefits of this technology are strong concealment, good directionality, and a long working distance. In practical, it can be difficult to find the appropriate statically positioned acoustic-excited targets in the sound field. In this work, the proposed method utilizes variational mode decomposition method to separate the characteristic modes of the multimodal vibration signals of moving targets in the sound field. Each mode is weighted, reconstructed, and merged to create the acoustically excited signal of the moving target based on the variations in amplitude and frequency bands between the moving object's macroscopic displacement signal and the vibration signal. Last but not least, the objective speech quality evaluation method of PESQ was used to assess the laser speech signal produced by the dynamic target laser speech acquisition method based on variational mode decomposition. According to the research, the standing person laser speech's PESQ score increased from 1.704 to 2.804, indicating a noticeable improvement in speech quality.

Steel structure has been widely used in modern buildings because of its excellent building performance such as lightness, low-cost. However, the poor fire resistant of steel structure can lead to a lot of hidden troubles including life safety and economic losses, cause the construction of steel structure become a key research direction in fire investigationwork. Currently, the experts for fire investigation have to analyse the surface trace of the steel evidence after the fire throughempirical data by person because of the lack of intelligent methods and equipment, which have a great influence on the investigation efficiency. This article proposes an intelligent detection method for steel structure fire traces based spectral imaging technology. The surface spectrum and morphology information of heated steel structure can be collected through standard spectral imaging equipment in 400-1000nm, then establishing the connections between the highest temperature or time and its feature of spectral imaging data. More than 150 groups of steel plate samples were prepared in various temperature including 100 ℃ to 1200 °C by muffle furnace and stacking fire in the experiment. The spectral imaging data can be obtained and imported to the classification and recognition algorithm, 90%of the samples conditions can be identified recognized accurately. The results indicate that spectral imaging technology can effectively assist in the development of fire investigation work by quickly and intelligently identifying fire traces, and has good application prospects in the field of fire investigation.

The refractive index is an intrinsic optical property of seawater and it is one of the essential parameters in ocean observation. The traditional method of detecting the refractive index of seawater is calculated by measuring the temperature, conductivity, and pressure of seawater using CTD (Conductivity-Temperature-Depth) device. However, CTD can only detect the conductive ions in seawater, but not the non-electrolytic solutes in seawater, which leads to errors in salinity detection, and the accuracy of current seawater refractive index detection methods cannot meet the requirements. In this paper, based on the requirements of seawater refractive index measurement, the Zemax optical simulation analysis of the seawater refractive index measurement system was carried out. The influence of the angle between the two pass lenses in the detection area, the focal length of the focusing lens, and the size of the photo sensitive surface of the detector on the measurement system are analyzed. When the angle between the two pass lenses is 30°, the focal length of the focusing lens is 300 mm and the size of the photosensitive surface of the detector is determined to 4×4 mm2, the optimal refractive index measurement sensitivity of the optical system is obtained. it is calculated that when the detector accuracy is 1μm, the refractive index detection sensitivity of the optical system is 10^-6 RIU.

The Flexray illumination technology is a crucial resolution enhancement technique for immersion photolithography machines operating at nodes of 28nm and below. The Flexray pupil shaping Module is the key component of this pupil shaping technique. This module enables the generation of arbitrary illumination modes by adjusting the angular spectrum of the beam through a micro mirror array (MMA). To meet the requirements of optical characteristic detection, a multiparameter detection method has been proposed. The optical performance parameters, including polar non-balance, opening angle, azimuth angle, and ellipticity, are simultaneously obtained by imaging the intensity distribution on the pupil plane. Experimental measurements and analysis of illumination shapes using the Flexray Pupil Shaping Module have been conducted. The experimental results demonstrate that this measurement method can be applied to test the performance of the Flexray Pupil Shaping Module in lithography system.

Micro-nano 3d measurement method based on fringe projection technology is widely used in many detection scenarios due to large non-contact, field of view, high accuracy and high adaptability. However, there is a problem that the nonlinear motion of the scanning actuator introduces random scanning error in the actual measurement system, which leads to the reduction of absolute accuracy and measurement repeatability. In this paper, a random scanning error suppression method is proposed to solve this problem. The principle of mutual calibration of the second order slope of the theoretical modulation curve and the second order slope of the actual modulation curve of the grating fringe is used to achieve the accurate value of the scan step. The feasibility and progressiveness of this method are proved by theoretical derivation, simulation and experimental verification, which demonstrate that the measurement repeatability is better than 1nm in simulation, the absolute precision is better than 3nm and the measurement repeatability is better than 2nm in experiment.

Multi-band terahertz metamaterial absorbers offer new perspectives to achieve perfect absorption and multipoint information matching, which enable an ever-growing number of applications. In this study, a dual-band terahertz metamaterial absorber based on the metal split ring is designed. The absorber has perfect absorption peaks at 1.15 THz and 2.47 THz, and the absorption rate is more than 99%. The absorber produces a harp peak with a bandwidth of 0.008 at 2.47 THz, which has an extremely high quality factor of 308. The distribution of the electric field and surface current at two resonance points is analyzed using the finite element integration method. Through full wave simulation calculation, the maximum sensitivity of the analyte refractive index of the absorber is 400 GHz/RIU, and the maximum sensitivity of the thickness is 35 GHz/μm. The results show that the absorber can achieve highly sensitive detection of trace substances.

As an important combat equipment for battlefield information acquisition, situation awareness and weapon guidance, photoelectric reconnaissance payload carries out the whole chain task of "detection-search-tracking-guiding-evaluation". In view of the large input delay of miss range, bad disturbance environment of the aircraft, target occlusion or large scale change of the aircraft, the miss range has wild value, and so on, it is necessary to suppress the tracking input delay and study the compensation delay prediction and compensation method of miss range. Firstly, the state prediction matrix and prediction error covariance matrix of the current time are constructed by the dynamic model of the moving target, the target acceleration value at the previous time is statistically analyzed and then fed back to the target state space model to update the current acceleration mean value, and the time-varying factors are corrected by the residual sequence, and the gain is adjusted adaptively, so as to improve the tracking performance when the state changes. The tracking simulation experiments on different maneuvering targets show that the tracking of maneuvering targets can be well realized with good tracking performance.

With the improvement of hardware computing power, the application of deep learning methods in the field of remote sensing is increasing. This paper summarizes the progress of deep learning methods in remote sensing image object detection in recent years. The main methods of deep learning methods to extract and use target feature information in various target detection tasks are summarized. Finally, the application trend of deep learning methods in the field of remote sensing image detection is prospected.

Distributed optical fiber acoustic sensing system (DAS) can achieve real-time monitoring of vibration events along the optical fibers by demodulation of back Rayleigh scattered light information in optical fibers. DAS system has a long detection distance, low cost, high sensitivity and other characteristics, and has been playing an important role in various engineering applications. However, in practical engineering detection of DAS system, there are other vibration sound sources in the environment besides target vibration events. However, the number and type of environmental sound sources are usually unknown. This kind of uncertain multi-source interference will have a serious impact on the monitoring of target events.

Aiming at the problem that the measurement of strong laser far-field focal spot based on side lobe beam diffraction inversion cannot extract the smallest measurable signal in the periphery of the side lobe image, this paper proposes a method for detecting the peak parameters of the weak signal region in the side lobe based on neighborhood vector principal component analysis (NVPCA) image enhancement. The main optimization measures are as follows: First, treat each pixel in the sidelobe image and its 8 neighborhood pixels as a column vector to construct a 9-dimensional data cube, and the first dimensional data after PCA transformation is an NVPCA image; Secondly, using angle transformation to transform the detection object, the various peak parameters of the one-dimensional sidelobe curve in all directions are detected, thereby obtaining the quantified energy distribution characteristics of the weak signal area of the sidelobe beam; Then, search for the maximum position points of each sidelobe peak in all directions, connect all position points to generate a maximum ring for each sidelobe peak, and calculate the grayscale mean of each maximum ring; Finally, the grayscale mean of the maximum rings that is greater than the LCM target separation threshold and the smallest is selected as the minimum measurable sidelobe peak signal of the entire sidelobe beam. The experimental results show that the sidelobe weak signal detection method based on NVPCA image enhancement can separate and extract the minimum measurable signal from the fifth peak ring on the periphery of the sidelobe image, increasing the dynamic range ratio by 1.528 times, improving the calculation accuracy of the dynamic range ratio, and laying a foundation for the future accurate measurement of strong laser far field in large scientific devices.

The invention of optical coherence tomography (OCT) provided a noninvasive high-resolution three-dimensional imaging technique for the visualization of the retina. To date, OCT has become the standard of care for daily ophthalmic clinical practice, especially for the posterior segment examination. Lateral resolution determines the smallest features that can be resolved by an OCT system on the lateral dimension and it is one of the most important parameters for OCT imaging quality evaluation. Therefore, the standardized testing device and method for OCT lateral resolution evaluation have become critical to certify its imaging and diagnosis quality. As a widely applied standard, ISO 16971:2015 published by the International Organization for Standardization specifies a test device and test method for the lateral resolution evaluation for OCT for the posterior segment of the human eye. In this work, we implemented the testing device and method described in ISO 16971:2015 with a commercial ophthalmic OCT instrument and evaluated its performance in lateral resolution measurement. Results show that the test device and method could provide a quick and coarse evaluation of the lateral resolution. There are improvements to be made so that the measurement can be more accurate, reliable, and consistent for manufacturers, users, and regulatory authorities to implement.

We propose a silicon based optofluidic one-dimensional photonic crystal biosensor structure for tumor marker detection, which is composed of a nanobeam resonator transducer with excellent detection limit performance, a filter with low sidelobe jitter and a microfluidics roof. Using the three-dimensional finite difference time domain method, a one-dimensional photonic crystal slot nanobeam resonator transducer optimization model consisting of a circular hole array linearly decreasing from the center to both ends was obtained. Under the influence of absorption loss of biological solution, the transducer works in the communication E-band, with the Q value still up to 10538, refractive index sensitivity of 338 nm/RIU, and refractive index detection limit of 10^-5 RIU, corresponding to the detection of fg/mL carcinoembryonic antigen, which can be directly used for the detection of tumor marker under the capture of antibody probes in microfluidics chip. By optimizing the apertures on both sides of one-dimensional photonic crystals with a taper shape, a cutoff filter with low sidelobe jitter can effectively filter out the high-order resonant peaks of the transducer, forming a large free wavelength detection range. The microfluidics channel is used to inject different refractive index liquids, and the cut-off wavelength can match the detection requirements of different concentrations of tumor markers. The sensor structure is expected to build a multi-channel parallel lab-on-chip through splitters and detect multiple tumor markers simultaneously.

High-resolution three-dimensional brain image reconstruction is crucial for understanding the brain. Light sheet microscopy combined with tissue clearing imaging plays a pivotal role in analyzing the micro-level structure of mammalian brains. However, the complex multi-level stitching process poses challenges such as non-overlapping areas, surface deformation, and tissue loss, resulting in incomplete or discontinuous tissue structures at the junctions. These issues not only impact the precision of the atlas but also complicate subsequent analyses like cell counting and neuron tracing. To address these issues, we propose a rapid deep learning-based image inpainting approach for accurate neuron reconstruction and analysis. Our approach involves initially employing conventional registration algorithms to preliminarily stitch brain sections together, followed by utilizing a neural network to predict and restore missing tissue with a thickness exceeding 10 µm. This process enhances the structural continuity and integrity between adjacent brain slices. Compared to the original 3D U-Net and ResNet models, our approach performs better and has a processing speed that is five times faster than the original 3D U-Net. Moreover, our method enables more accurate cell counting by repairing incomplete cell bodies, leading to an average improvement of 37.37% in the number of cell bodies accurately counted near the slice junction. By integrating this novel 3D image inpainting network into brain reconstruction processes, our research opens new avenues for a more detailed and accurate investigation of neural circuitry and neurological disorders.

Traditionally, diffusion correlation spectroscopy (DCS) calculates blood flow (BF) by measuring the temporal intensity fluctuations of multiple scattered light by DCS devices that fit all the requirements, which consists of a source-detector pair, a data acquisition card, and a host computer. In this paper, a portable DCS device based on STM32 microcontroller unit (MCU) is proposed in order to solve the drawbacks of traditional DCS devices, which are costly and bulky. Arterial occlusion experiment was established by monitoring the subject's arm blood flow changes based on two devices, simultaneously. The experimental results indicated that excellent agreement of blood flow changes measured and intensity auto-correlation function detected by portable DCS device and traditional DCS device.

Currently, vocal cord lesion diagnosis of laryngoscopic images mainly relies on physicians' expertise and clinical experience. This greatly increases the work pressure of physicians and has limited efficiency. To solve the above problems, this study aims to construct a deep network structure named VCLR-Net based on the improved YOLOv7 to achieve the detection and recognition of vocal cord lesions. First, Convolutional Block Attention Modules (CBAM) are added to the HEAD network to improve the focus of color and spatial features on lesions. Next, the Alpha Intersection over Union loss (AlphaIOU) loss function is used to improve the robustness of the lesion recognition model. In the experimental results, the proposed VCLR-Net network achieves mAP and F1 of 0.762 and 0.748 in the image dataset. The network enables accurate lesion recognition for a large number of laryngoscopic images.

Current research is focused on the miniaturization and integration of detection devices for biological targets in order to adapt to outdoor and emergency field settings, as well as to enhance the flexibility and practicality of detection. This study proposes an ultra-high integrated silicon nitride fluorescence excitation chip for point-to-point excitation of biological fluorescence signals. By employing the finite-difference time-domain(FDTD) method and utilizing silicon nitride as the functional material, a visible light fluorescence excitation chip operating at the wavelength of 645nm was designed and fabricated. The chip is composed of an input grating coupler, multiple multimode interferometers, and output grating couplers. A grating array containing 2016 excitation points is constructed cascading multiple multimode interference splitters and connecting excitation gratings at the terminals. The single grating coupling efficiency of the chip is 30% with a total excitation light area of 4×4mm² and an emitted light angle of around 80 degrees. By adjusting the positions of different output grating couplers, the chip can adapt to excitation requirements at different locations, making it suitable for various applications such as fluorescence quantitative polymerase chain reaction(PCR) and digital PCR. With the help of microfluidic chambers and a photodetector, the chip successfully achieved the detection limit of 0.5 µmol/L Cyanine 5(Cy5) fluorescent reagent solution, which is sufficient for direct detection of PCR fluorescence signals.

The relationship between neural activity, brain function, and corresponding biological behaviors remains a significant challenge in neuroscience. Exploring this relationship needs various optical imaging techniques to acquire real-time data with high spatial resolution. A promising technology recently is wearable miniaturized microscopes (mini scopes), which enable long-term neural activity recording in freely moving animals. However, most one-photon mini-scopes have limitations for imaging in-depth with high resolution and large field of view (FOV). To address this, we developed a one-photon miniaturized fluorescence microscope (1P-miniFM), intended for imaging of live brain neurons in free-behaving animals at subcellular level (~1.2 μm). We conducted specially designed optical path, achieving an imaging FOV of ~700 × 400 μm. In addition, we incorporated an electrowetting lens (EWL) to achieve a wide range of ~300 μm z-axis scanning with little resolution loss. 1P-miniFM is compact (11 × 17 × 24 mm) and lightweight (~2.9 g), causing little impediment to animals’ spontaneous behaviors. With genetically encoded calcium indicator GCaMP6s, we monitored neuron activities in secondary motor cortex (M2) during consecutive pain-related and sensory stimulations. We found that M2 neurons are key components and exhibit distinct variations in the response patterns. 1P-miniFM has potential as an excellent tool to explore relationships between neuron network and animal behaviors.

As an optical biopsy method, PZT fiber-optic scanning two-photon endomicroscopy holds significant importance for the early detection of gastrointestinal tumors. The scanning attributes inherent to the PZT fiber scanner wield direct influence over the electro-mechanical features of the two-photon endomicroscopy, rendering it a pivotal component of distal-scanning endoscopes. This study delves into a comprehensive analysis, both in the modal and frequency domains, of a forward-fixed PZT scanner from a theoretical standpoint. Additionally, we fabricated a forward-fixed PZT fiber scanner and validated its first- and second-order frequency domain characteristics experimentally. Theoretical simulations and experimental findings demonstrate that the forward-fixed scanner exhibits superior driving capability compared to the reverse-fixed scanner.

An quantum access network, connecting multiple end users and a backbone network in a scalable and cost-effective way with tree or star structures, is suitable to combine with continuous-variable quantum key distribution for network security. Under the realistic condition, in a passive optical network in optical distribution network achieved by a practical passive beam splitter, it is hard to divide the signal into 1/n parts evenly. To well quantify the impact of the imperfection, we modify the formula for secret key rate calculation with the imperfection effect and conduct numerical simulations. The results indicate that if the communication parties neglect this imperfection, Eve can attack the entire network and attain information without the parties’ awareness.

With the continuous development of optical quantum sensor technology, the importance of quantum entanglement sources is growing. As an essential component of optical quantum systems, entanglement sources have broad applications in quantum communication, imaging, computing, and other fields. However, the high-power consumption and preparation costs associated with these sources hinder their integration and expansion. To address these issues, we conducted experiments using a 40mW laser diode and a typical type-II barium metaborate (BBO) crystal to prepare a low-power and low-cost quantum entanglement source. Our results show an average count of over 15,000 entangled photons per second and over 7,500 entangled photon pairs per second, which are in good agreement with theoretical analysis. We have made significant progress towards achieving the goal of low-power and low-cost entanglement sources, which will facilitate the practical application of quantum entanglement lighting and promote the wider adoption of optical quantum systems in the future.

The track before detect technology based on Hough transform is an effective method for hypersonic target detection, but this method requires the target to move as linearly as possible. In practice, hypersonic targets often generate significant maneuvers during flight, such as “snake maneuver” where the accumulated values from the previous time will have a notable impact on the accumulated values at the current time. Aiming at this issue, this paper proposes a recursive accumulation method based on attenuation factor. By setting an attenuation factor with a value less than 1 and using time difference as the power of the attenuation factor, the accumulated value at the previous time is weakened, ensuring that the accumulated value at the previous time can be accumulated while reducing its impact on the current accumulation. At the same time, it avoids the computational burden caused by the continuous deletion and addition of measurement point information in the original recursive method, which has certain application prospects.

In order to solve the problem of cooperative pose estimation in augmented reality systems, a newcooperative poseestimation method is proposed. Firstly, statistical analysis of the neighborhood of point cloud space is carriedout toimprove the precision of high point cloud segmentation. After point cloud processing, the collaborative pose estimationmodel is established by using centroid to obtain the collaborative pose estimation results. Experimental results showthat the proposed method can effectively obtain collaborative pose estimation results.