Fiber optic Raman biochemical sensor is a fiber optic sensor based on the principle of Raman scattering.It combines the advantages of Raman scattering spectroscopy technology and fiber optic transmission and has the characteristics of high sensitivity, strong specificity, small size, real-time speed, and unlabeled detection. It can be widely used in the detection of low-concentration analytes in the field of biochemical detection. The basic principle,structure, and working mode of optical fiber Raman biosensors are introduced. Starting from its practical applications in food safety, environmental monitoring, biomedicine, and other fields, the preparation process of optical fiber Raman probes with different structures and the detection limits of actual analytes are emphatically introduced. In addition, the current challenges and future development directions of fiber optic Raman biosensors were summarized, and further discussions were conducted on how to improve the performance of fiber optic Raman sensors.

In recent years, the application of the Tyndall effect, with its immediate response and non-invasive nature,has been widely used in various fields, including environmental biology, medical diagnosis, chemical production,food, and drug. This paper briefly reviews existing techniques for applying the Tyndall effect, specifically, the application of the Tyndall effect in different scenarios, such as heavy metal ion concentration detection, biomolecule concentration detection, disease diagnosis, and preparation. Additionally, its applications in industrial production for auxiliary hydrophobic materials and semiconductor chip production, improving the energy efficiency of solar cells, and food safety for white wine quality identification and tea pesticide residue detection are discussed. Finally, the future directions for applying the Tyndall effect are presented. The summary and outlook of applying the Tyndall effect will provide research directions and theoretical references for subsequent applicators.

at Aug. 8,2021, the experiment at the National Ignition Facility (NIF) made a significant step toward ignition, achieving a yield of more than 1. 3 megajoules (MJ). This advancement puts researchers at the threshold of fusion ignition, an important goal of NIF, and opens access to a new experimental regime. The record-setting experiment was the culmination of years of research and development in lasers, optics, diagnostics, target fabrication, experimental design, and computer modeling and simulation. In this article, we have investigated the development of fusion ignition in NIF, to provide the reference for Chinese laser fusion research.

Quantum Interference Radar (QIR) is a new radar regime based on quantum mechanics and exploits the quantum entanglement effect for target detection, and the variation in the number of photons detected by QIR has an impact on the radar resolution and system survival function. In order to reduce the influence of the liquid water content of stratocumulus on the detection performance of the QIR system, the optimal average photon number adaptive algorithm(Photon Number Adaptive Algorithm, PNA) based on the decoy state quantum key protocol is proposed. The relationship between the liquid water content of the stratocumulus, the transmission distance and the optimal average photon number is established, and the QIR resolution and the system survival function before and after the adaptive adjustment are compared. The theoretical analysis and simulation results show that when the pulse wavelength is fixed,the stratocumulus particle concentration is 1. 579×104 cm-3, and the stratocumulus liquid water content is 0. 5 g/ cm3,the QIR angular resolution decreases from 1. 289 to 1. 028 after adopting PNA strategy, and both the angular resolution and spatial resolution of the QIR system are effectively improved. when the liquid water content of stratocumulus is 0. 103 4 g/ cm3, the channel survival function improves from 0. 229 2 to 0. 416 7. Therefore, the reliability of QIR detection in stratocumulus can be improved by adaptively adjusting the average number of photons contained in the signal pulses of the sending end of the system through PNA strategy.

To meet the light source requirement for image positioning of the lightning imager on board the Fengyun-4 meteorological satellite, a laser with a central wavelength of 777. 4 nm and a continuous output power of 1. 5 W is designed based on a two-stage master oscillation power amplifier structure (MOPA) and a periodically polarized lithium niobate crystal doped with magnesium oxide (MgO: PPLN) to double the frequency. Using 940 nm off-peak pumping technology to suppress the amplified spontaneous emission (ASE) and stimulated Brillouin scattering (SBS) in the optical amplifier, we obtain a single-frequency narrow linewidth laser with an output power of 15 W, a signal-tonoise ratio of 59. 6 dB, a linewidth of 323. 4 kHz, and a relative intensity noise of -145 dBc/ Hz. The optical-optical conversion efficiency of the amplifier is up to 37. 2%. Based on the theoretical analysis of the doubling efficiency, temperature-tuned bandwidth, and acceptable wavelength bandwidth of the MgO: PPLN crystal used, 777. 4 nm continuous light with a power of 1. 5 W is obtained using a single-pass doubling method.

With the development of laser propulsion, laser photon propulsion has become the focus of propulsion technology for deep space exploration. At present, laser photon propulsion is mainly based on continuous laser with long-distance transmission irradiation. Short pulsed laser has the advantage of high peak power. Therefore, the nanosecond high-repetition-frequency pulse laser for photon propulsion technology is proposed in this paper. In this paper,the impulse characteristics of photon propulsion with high-repetition-frequency nanosecond pulsed laser are studied by using the torsion pendulum. In the experiment, the laser wavelength, the pulsed width, and the repetition frequency is 1064 nm, 160 ns, and 10 kHz, respectively. The reflectivity of the output coupling mirror is 90%. The corresponding laser power range in the cavity is from 144 W to 414 W in the experiment. The local sliding fitting method is used to realize the smooth and noise reduction of the original data. The experimental results show that the impulse increases linearly with the increase of the output laser power. The impulse and the average thrust during the pulse time are calculated,respectively. The research results of this paper provide technical reference for the development of laser photon propulsion in space propulsion.

A novel optoelectronic voltage sensor together with its compensation method for temperature drift of sensing signal is proposed which is based on voltage-to-frequency conversion. The voltage sensing system is mainly composed of resistance voltage divider, voltage-to-frequency converter, light emitting diode, optical fiber, photo detector,and sensing signal processor. Two pieces of AD654 chip are respectively used as voltage and temperature sensors.The real-time compensation for temperature drift of voltage sensing signal can be achieved by using the nonlinear fitting for experimental data from two branches of light sensing signals. The dc voltage in the range of 0. 6~5. 0 kV has been measured by use of the above proposed voltage sensor. The temperature drift of voltage sensing signal has been suppressed effectively. And the average relative error due to temperature drift is within ±0. 7% in the temperature range of -40~60 ℃.

We propose and numerically demonstrate that short cavity semiconductor laser with dual-optical-feedback can stably generate broadband chaotic signal. Through theoretical analysis and comparison of the effects of the feedback factors on the dynamic characteristics of the laser with single-optical-feedback and dual-optical-feedback structure, it was found that the short cavity semiconductor laser with dual-optical-feedback can generate chaotic signals within a large parameters range. The analysis of the signal characteristics of different output states of the laser shows that the laser with dual-optical-feedback structure evolves through a period-doubling bifurcation leading to chaos.In addition, the influence of the feedback factors of the dual cavity on the bandwidth of the output signal was studied,and it was identified that the bandwidth of the chaotic signal generated by the laser with dual-optical-feedback structure can reach over 10 GHz, and gradually increases with the increase of the feedback factors.

In order to meet the need of high pointing accuracy of fast steering mirror in photoelectric system, the two-dimensional Angle calibration technology of fast steering mirror based on eddy current sensor is studied. The working principle of two-dimensional Angle measurement using eddy current sensor is analyzed, and a fast steering mirror Angle calibration system is designed. Two calibration models, polynomial fitting and bilinear interpolation, are analyzed theoretically and tested experimentally. The results show that the calibration error of the bilinear interpolation method is within 2″ in the angle range of ±1 000″, which is higher than the traditional calibration accuracy of the polynomial fitting method. The improvement of calibration accuracy is of great significance for realizing high precision pointing of fast steering mirror.

In large electrical equipment, cables play an important role in transportation and signal transmission,and measuring the bending radius of cables is an important link in ensuring the quality of the installation process. A non-contact machine vision cable bending radius measurement method is proposed to address the issues of inconvenient operation and low efficiency in traditional methods for measuring cable bending radius. Based on deep learning RGBD bimodal semantic segmentation network, cable masks are segmented, and cable feature centerlines are extracted from mask images. Then, a cable spatial feature point set is constructed, and the cable spatial feature curve is reconstructed using parameter curve algebra to reconstruct the cable spatial feature curve, Then, the cable bending radius is calculated by using the Curvilinear motion law of the particle space. To verify the practical application effect of the method proposed in this paper, a standard arc with a radius of R = 110 mm and 125 mm was used to fix the cable.The comparison between the measurement results of the manual ruler using the chord height method and the measurement results of the method proposed in this paper shows that the average error of measuring the bending radius using this method is reduced by 7. 2% and 2. 2%, respectively, compared to the manual ruler measurement, and the measurement time is reduced by 93. 8% and 92. 2%, respectively. The proposed non-contact machine vision cable bending radius measurement method algebraically reconstructs the cable spatial characteristic curve based on mask images,calculates its spatial bending radius, and the measurement results are reliable, with higher accuracy and faster speed than manual measurement. It can be applied to various occasions of cable bending radius measurement tasks.

A technique of fast ellipsometric measurements based on dual-drive elastic modulation is proposed in paper, for which the measurement speed of traditional mechanical rotary compensator ellipsometer is slow. The modulation mode of 45°dual-drive photoelastic modulator (PEM) working in pure traveling wave mode was studied by theoretical analysis and simulation. The results show that the fast axis direction can high-speed periodic rotation and the phase delay is constant when the PEM working in pure traveling wave mode. Combined with the principle of ellipsometry,a theoretical model of fast ellipsometric measurement based on dual drive PEM was studied in detail. The results show that the optical period of this model was less than 10μs, and the ultra-high speed optical modulation could be realized.With this study, it is expected to solve the problem of ultra-high speed detection of thin films in the production process of semiconductor integrated circuits, microelectronics and photovoltaic cells. It is of great significance and academic value for in-line measurement of thin films.

Due to the temperature sensitivity of semiconductor materials, the photovoltaic response output characteristics of photovoltaic cells will change with the temperature variations caused by laser irradiation. This paper focuses on the interference of repetition frequency laser irradiation on the response characteristics of silicon-based photovoltaic cells. Starting from the principle of photovoltatic effect, based on the response output model and one-dimensional heat conduction equation, the effects of different pulse parameters on the voltage output of photovoltaic cells under laser irradiation are calculated. The results indicate that repetition frequency lasers can significantly impact the normal operation of photovoltaic cells. However, by controlling the pulse energy and selecting appropriate parameters, such as pulse width and pulse interval, significant interference effects can be achieved.

Aiming at the defect that the traditional gray gravity method cannot accurately extract the center line of the region with large curvature change in the line structured light strip, an algorithm for extracting the center line of line structured light based on improved gray barycenter method is proposed. The region of interest is extracted based on the threshold method, and the noise points in images are filtered through the open and close operation to obtain the image to be processed with relatively high quality. Center points of the light strip through the gray gravity method. And a rectangular calculation area is set up. The direction vector and normal vector of the initial points are extracted according to the least square method. Finally, the offset of the initial point in the normal direction is recalculated in the rectangular area to obtain the exact coordinates of the center point. Experimental results show that the proposed algorithm can accurately extract the center line from the region with large curvature change in the center of line structured light,achieving a mean precision of 0. 096 pixels, a mean precision rate of 44. 2%, and an average extraction speed of 0.082 seconds.

A concrete crack detection method based on improved YOLOv4 is proposed to address the problems of low detection accuracy, large number of model parameters and slow detection speed in current deep learning methods for detecting concrete cracks. Firstly, the backbone feature extraction network of YOLOv4 is replaced by the lightweight network Mobilenetv1, and the ordinary standard convolution in the enhanced feature extraction network of YOLOv4 is modified into a depth-separable convolution; secondly, the lightweight attention module CBAM (Convolutional Block Attention Module) in the PANet module to improve the accuracy of crack target detection with a controlled amount of parameters; finally, the Spatial Pyramid Pooling (SPP) module in YOLOv4 is replaced by the RFB-s module that simulates human vision. The experimental results show that compared to conventional YOLOv4, the mAP of this model increases by three percentage points, the amount of parameters is reduced to 14 M and the detection speed is up to 42 frames per second.

Prior to the correlation filter target tracking algorithm, the research work on low-illumination widefield tiny target tracking has not been reported. In response to this, this paper proposes an algorithm that combines an image difference detection framework with a correlation filter (CF)-based tracking framework, and introduces dual filters to combat the adverse factors brought by the environment. A sparse response regularization term of the l2 norm is proposed to suppress the abnormal peaks produced by the CF framework. In the response phase, the positions of small objects are predicted based on the dual filter weighted fusion. The results show that the proposed algorithm has excellent tracking performance for fast moving, deformation and motion blur of small targets in low illumination and wide field of view, and meets real-time performance. A new dataset of 41 night surveillance sequences captured by Eagle Eye cameras was collected as a benchmark. The experimental results show that the algorithm in this paper improves DP by 8. 8%, AUC by 7. 4%, and realizes real-time operation of 30. 6 frames per second on a single CPU.

Accurate identification and positioning of tank targets is an important research in information warfare.Aiming at the problems of insufficient timeliness and low accuracy of traditional detection algorithms, a detection algorithm based on improved YOLOv5 tank automatic identification was proposed. The YOLOv5 model is used to identify tank targets in complex battlefield environment. The Attention-based information fusion module is introduced into the basic model of YOLOv5 to improve the detection accuracy and identification ability of the model. The Pre-segment multi-scale fusion module is used to solve the problem of information loss caused by pooling operations in backbone network. Use Swin Transformer to reduce the leakage rate of small target tanks. The experimental results show that compared with the original YOLOv5 model, the accuracy rate, recall rate and average accuracy of the improved model are increased by 0. 9%, 11% and 5. 7%, respectively. The improved YOLOv5 model can accurately identify tank targets in a complex environment with a large field of vision, reducing the problem of tank small targets missing detection.

Eccentric photographic vision screening equipment is an important means of rapid detection of refractive state, and pupil image segmentation is an important part of its imaging algorithm. Aiming at the problems of limited computing resources and low precision of pupil segmentation in embedded devices, a lightweight pupil image segmentation algorithm based on improved Mobile-UNet was proposed. Based on U-Net improvement, the algorithm is preliminarily lightweight by using inverse residual linear bottleneck module. Group convolution is used to reduce parameters,channel mixing is used to open inter-group channels, and an adaptive parameter fusion parallel attention mechanism is introduced to improve segmentation performance. In addition, the optimization of the loss function enhances the attention to the boundary. The experimental results show that compared with MobilenetV2, the number of model parameters is reduced by 90%, the number of floating point operations is increased by 19%, but the segmentation performance is significantly improved. Compared with U-Net, the complexity of the model is greatly reduced and the segmentation performance is improved. Compared with other algorithms, this model has advantages in complexity and segmentation performance, and achieves lightweight and efficient segmentation.

In order to improve the low light image enhancement algorithm based on convolutional neural network (CycleGAN, Retinex-Net, etc. ), which has the problems of excessive model parameters, high memory consumption and poor image recovery quality, we propose the low light image enhancement algorithm HBTNet incorporating the half-wave attention module based on the lightweight algorithm IAT. In order to improve the spatial information loss caused by frequent convolution of the network, the half-wave attention module is introduced into the network, which can effectively obtain the characteristics of wavelet domain, enrich the contextual information and improve the feature extraction ability. The quality of image recovery is improved by introducing MS-SSIM loss function used to preserve the edge and detail information of images. The experimental results show that HBTNet improves PSNR by 2. 69% and SSIM by 5. 56% compared with IAT algorithm on LOL dataset. the number of model parameters of HBTNet algorithm is only 0. 11 M, which can meet the real-time requirements of end users.

A temporal feature based image registration algorithm is proposed, which achieves effective registration of visible light and infrared non fixed dynamic target images through the analysis and extraction of temporal grayscale features. The focus was on analyzing the anti noise ability of the algorithm. When noise has a significant impact on registration accuracy, a preprocessing filtering method was designed to reduce the impact of noise on temporal feature extraction,thereby reducing the algorithm's mismatch rate. Using candle flames as the detection target, experiments have shown that the algorithm has good noise resistance at an image signal-to-noise ratio of 10 dB, and can achieve visible and infrared image registration of non fixed targets. When the signal-to-noise ratio of the target image is equal to 6dB, the algorithm cannot complete image registration. In this case, the preprocessing filtering method is used to reduce the mismatch rate of the algorithm and successfully achieve visible and infrared image registration of non fixed targets.The key parameter of the preprocessing method calculated in the experiment is 0. 25, which has the best effect.

This paper proposes a technique for enhancing weak smoke information at the beginning of a fire by using the information between video image sequences. An improved ViBe algorithm is used to monitor the entire video and obtain the motion mask of each frame. The smoke information is amplified by utilizing the frame difference amplification algorithm that removes outliers. Moreover, improving the contrast of smoke regions through the use of an improved CLAHE algorithm. Smoke information is preserved using a merge mask, while the flip mask retains background image information. Finally, the smoke area and background area are merged to obtain the final image. Experimental results show that the algorithm can significantly improve the contrast of the smoke leaking area. As the original image background is fused, the noise is significantly suppressed.

Aiming at the problems of long-time, high-energy consumption and missing information for labyrinth robot to obtain labyrinth information independently, this paper proposes to use image processing related methods to obtain labyrinth cell information, so as to improve the labyrinth task efficiency of labyrinth robot. Aiming at the inherent defects of uneven brightness and color dissimilation in labyrinth images, a labyrinth cell edge detection method based on Kirsch and improved Canny operator is proposed, which realizes the edge feature extraction of labyrinth cells. Using the principle of image morphology, the labyrinth cell features are denoised, merged, expanded and filled to achieve image enhancement of the labyrinth cell. Aiming at the characteristics of labyrinth cells, a pixel block matching feature method is proposed to match the information of labyrinth cells to achieve the information extraction of labyrinth cells.The experimental results show that the accuracy of information extraction for labyrinth images with a brightness range of γ =1. 1 and a shooting angle of 30 degrees or more reaches 100%. Therefore, the labyrinth information extraction method proposed in this paper is practical and feasible, which provides a guarantee for the efficient execution of the subsequent labyrinth tasks of the labyrinth robot.

In the process of image segmentation, if the noise in the image cannot be effectively suppressed, the segmentation accuracy of the image will be directly affected. In order to improve the segmentation effect of the image,a light change image segmentation method based on laser vision is proposed. The image is input into the laser vision system for noise removal, target enhancement and other processing. During the process, the control sensor in the system closely tracks the image processing process to improve the image processing effect; Obtain the image histogram,determine the selection range of the gray histogram threshold of the light transformation image according to the smoothing result of the histogram, divide the image into regions to obtain the corresponding threshold, and complete the accurate segmentation of the light image according to the determined threshold. The experimental results show that this method can effectively remove the image noise and achieve a good segmentation effect, it can achieve efficient and accurate segmentation of image targets.

In order to improve the effect of infrared image super-resolution reconstruction, an infrared image super-resolution reconstruction method based on the fusion of visible light and near-infrared light based on depth learning is proposed. The salient region detection model of infrared image is established by using the reflective characteristics and infrared radiation characteristics of infrared image; The edge contour features of the image are detected by the appearance difference level between the visible light and near-infrared images, and the fusion feature parameters of visible light and near-infrared light are extracted; According to different fusion levels, image signal level, pixel level, feature level and decision level are reconstructed to extract image edge, shape and texture features; According to the noise level of the feature distribution and the registration quality, the infrared image super-resolution reconstruction is realized by using the depth learning algorithm. The simulation test results show that the method has a strong ability to detect the salient features of infrared image reconstruction, and the image resolution is improved to 1 280×960 PPI, the template matching accuracy is 49. 4%, the peak signal to noise ratio PSNR value is higher than 36. 34 dB, and the structure similarity SSIM value is higher than 0. 972. The reconstruction effect is good, and it is more suitable for infrared image target feature recognition in specific scenes.

Aiming at the problems of index modulation orthogonal frequency-division multiplexing technology with high peak-to-average ratio, sensitivity to phase noise and carrier frequency, and high system complexity, this paper proposes a visible OFDM-IM system based on wavelet boost transform. Firstly, the orthogonal wavelet group is selected as the subcarrier, and the signal is divided into high-frequency signal and low-frequency signal through the splitting,prediction and update of the signal, and then combined with the visible light channel to form the LWT-OFDM-IM system,and finally the reliability, peak-to-average ratio characteristics and optimal wavelet decomposition layer of the system are simulated and verified by theoretical analysis and Monte Carlo method, and the results show that when the subcarrier N=256, the number of neutron carriers in the subblock L= 4, the number of activated subcarriers k = 2,and the system bit error rate is 10-4. LWT-OFDM-IM is about 8 dB better than FFT-OFDM-IM and about 4 dB higher than DWT-OFDM-IM. When the system complementary cumulative distribution function is 10-1 orders of magnitude,the peak-average improvement of LWT-OFDM-IM is about 2. 3 dB compared with FFT-OFDM-IM. As the number of wavelet layers increases, the better the bit-error performance of the system, when the wavelet layer is 3 layers,it is about 10 dB, and the bit error rate can reach the order of 10-5.

Common-signal-driven synchronization in two semiconductor lasers has been proven to achieve secure key distribution, but a high-level correlation between driving and response signal reduces the security. In this paper,we demonstrate a secure key distribution scheme based on a transformation module to enhance the protection efficiency of chaos synchronization-based secure key distribution. The chaotic signal is subjected to a delayed interference by the Mach-Zehnder interferometers (MZI) to eliminate the residual correlation in intensity. Subsequently, the final keys were generated by sampling and quantizing the chaotic waveforms from MZI utilizing the dual-threshold method. A proof experiment for secure key distribution at the rate of 0. 23 Gb/ s through 100 km fiber transmission was implemented.The experimental results prove that the transformation module introduced in the chaos synchronization-based key distribution scheme can effectively prevent an eavesdropper extracting keys directly by intercepting the output waveforms from the intensity variation of the driving source and realize the secure key distribution for legitimate users.

Indoor visible light localization has high requirements in terms of accuracy, in order to solve this problem,a Sparrow Search Algorithm (SSA) is proposed to optimize the indoor visible light fingerprint localization algorithm of Deep Belief Network (DBN). Firstly, the signal strength characteristic value and position coordinates are used to establish an offline fingerprint database. Secondly, the good global exploration and local development capabilities of the sparrow search algorithm are used to optimize the initial weight threshold of the deep confidence network, establish a network training model, and predict the position of the positioning target, so as to avoid the problem of DBN falling into local optimization and slow convergence. Finally, using the established offline fingerprint database data,the positioning error is calculated and analyzed. In the space experiment, the results show that the average positioning error of the algorithm in the paper is 3. 51 cm, and the probability of the positioning error within 6cm is 89. 9%, which is about 22. 5% lower than that of the DBN positioning algorithm.

Aiming at the problem that it is difficult to balance the detection accuracy and real-time in the current vehicle type recognition process, an improved road vehicle type recognition network based on RepVGG-A0 is proposed,which uses the idea of structural re-parameterization to fuse the multi-branch network to improve the network reasoning speed. The mixed void convolution is used to replace the traditional convolution, which strengthens the recognition ability of the model for large targets. Integrating the residual structure coordinate attention (RES-CA) module into the network backbone improves the network's ability to extract effective feature information, and avoids the impact of gradient disappearance and gradient degradation. In addition, the label smoothing regularization method is used to improve the loss function, reduce the impact of model overfitting on the detection results, and improve the generalization of the model. After verification, the recognition accuracy of the method in this paper on the road vehicle data set BIT-Vehicle has reached 97. 17%, which is 2. 67% higher than the original model, and is superior to the existing ResNet-18, VGG and other network models, while ensuring the detection speed of the model.

Aiming at the problem that the calculation of the average hop distance in the prototype DV-Hop localization algorithm produces large localization errors, this paper proposes a CADV-Hop algorithm for classifying the average hop distance. First, the inconsistent distribution of single-hop distance for different hop-count paths in wireless sensing networks is revealed, and the reasons and laws for the appearance of such differences are analyzed. Secondly,the classification average hop distance is calculated by classifying the inter-beacon paths according to the hop count and then calculating the classification average hop distance for each type of paths. Finally, on the basis of the more accurate distance estimation between the unknown node and the nearest beacon node provided by the classification average hop distance, the node coordinates are finally solved by combining the weighted least squares method. Simulation experiments show that the CADV-Hop algorithm effectively reduces the localization error without increasing the complexity of the algorithm as well as additional hardware, and with the increase of the number of beacon nodes, the CADV-Hop algorithm has higher localization accuracy than the prototype DV-Hop algorithm and the two improved DV-Hop algorithms.

In order to improve the stability of network transmission, an automatic identification algorithm for abnormal traffic in elastic optical networks based on isolated forest algorithm is proposed. Perform spectral density detection based on the abnormal distribution characteristics of traffic and the differences in normal data, construct a spectral feature extraction model for elastic optical network traffic, implement spectral feature filtering for abnormal traffic through low-pass filter convolution vector reorganization, adopt isolated forest algorithm to achieve adaptive optimization control for network traffic anomaly detection, and combine multi-dimensional spatial structure reorganization method to achieve detection and recognition of abnormal traffic in elastic optical network. The results showed that the missed detection rate and the false detection rate were relatively low, 3. 16% and 1. 03%, respectively. The detection takes less time, only 16 seconds. During detection, the external intrusion rate does not exceed 1%, and the immunity is strong.

Aiming at the problems of poor coordination and low accuracy in tracking angle and speed based on a given reference trajectory in the conventional unmanned driving trajectory tracking process, a laser sensor based unmanned driving transformation trajectory tracking system is proposed. The hardware structure including laser sensor,motor control, LCD display, host computer communication module, keyboard control, and microcontroller is designed.The angular velocity co controller error is adjusted based on the angular velocity and velocity error, and the offset number is weighted by neurons to optimize the dual push angular velocity co control effect; By introducing a sliding mode controller, the obtained speed control commands and angular velocity control commands are returned to the angular velocity co controller. Through the design of the trajectory controller and angular velocity co controller, unmanned transition trajectory tracking control is achieved. The experimental results show that after applying this system, the angular trajectory tracking results are closer to the desired angle, and the trajectory tracking error in both linear and curved states is smaller, with a maximum of 4 cm. The trajectory tracking accuracy is high, the positioning accuracy is accurate,and the overall application effect is better.

Aiming at the problems of misassociation, track interruption and poor adaptability in traditional 3D multi-object tracking algorithms in complex scenes, corresponding improvements is made in the data association stage,and a multi-object tracking algorithm based on weighted aggregation association cost and the prediction confidence of objects is proposed. Firstly, the weighted aggregation association cost is calculated by combining the location, appearance,and orientation features of the object to measure the difference between the objects. Then, the related conception of prediction confidence is introduced in the association cost matrix, and the association search domain of the missing target is adjusted according to the confidence. Finally, Kalman filter is used to update the object motion state and the prediction confidence. Experimental results on the measured data show that the proposed algorithm can improve tracking accuracy in the case of point cloud occlusion and trajectory intersection, and the MOTA reaches 73. 6%.

Laser cleaning is a new " green" paint removal technology developing rapidly in recent years. To address the problems of low efficiency and environmental unfriendliness in removing the surface paint of carbon fiber reinforced composites (CFRP) by chemical treatment and manual polishing, this work studied the regulation law and the removal mechanisms of CFRP surface paint cleaning by CO2 laser. The parameters such as the number of scans, beam overlap rate, scan speed, and laser power were studied on the cleaning behavior of the paint with a thickness of about 100 μm, and relatively optimized treatment parameters were obtained. The results indicated that when the beam overlap rate was 95. 75%, the scanning speed was 170 mm/ s, and the laser power was 44 W, the paint can be effectively removed in one scanning treatment with minimal damage to the fibers. The results provide a reference for further research removal of different thicknesses and types of CFRP surfaces.

Aiming at the problem of poor recognition accuracy of low slow small target in laser defense, a method of low slow small target recognition in laser defense based on embedded computer is proposed. This research first analyzes the principle of laser radar target detection and recognition, completes the architecture design of laser defense system based on embedded computer, and then completes the recognition design of low slow small targets for laser defense through target local feature extraction method and target recognition detection method. Finally, simulation experiments are carried out to prove the application effect of the proposed method. The experimental results show that the recognition accuracy of the proposed method for low, small and slow targets is higher than 91. 5%, and the average feature extraction time is less than 11. 3 ms, which is better than the comparison method.

In order to solve the problem that the calibration of fisheye camera with automatic parking vehicle is time-consuming, an improved chessboard corner detection method based on growth is proposed in this paper. On the basis of the original algorithm, the proposed hybrid distortion model can quickly determine the two initial directions of corners in different regions, so as to quickly improve the speed of rough positioning. At the same time, according to the particularity of production line calibration, the threshold of double standard deviation is proposed as the criterion to judge the initial chessboard. The experimental results in the production line environment show that the algorithm can not only meet the accuracy requirements of calibration, but also greatly improve the detection speed, and has a certain practical value for the calibration of production line camera.

Due to the space limitation, the indoor mobile robot requires higher confirmation of position during operation,and the problem of village moving track positioning deviation. Therefore, a dynamic localization method of indoor mobile robot based on embedded laser radar is proposed. The collected lidar data are preprocessed by coordinate transformation and European range segmentation to remove the influence of noise on subsequent positioning. The echo intensity is used to determine the position fitting change of the reflector, and the trilateral positioning algorithm is used to precisely locate the mobile robot. The experimental results show that the positioning error of this method is small. It can not only realize the high-precision dynamic positioning of the indoor mobile robot as a whole, but also monitor the trajectory of the indoor mobile robot very well.

In order to complete the drive control of electric vehicle better, a drive control method for electric vehicle with high resolution laser encoder is proposed. The high-resolution laser encoder is used to collect the operation related input signals of the electric vehicle drive system. The improved particle swarm optimization algorithm is used to preprocess all signals, identify all undetermined parameters in the model, and correct the amplitude deviation and orthogonality deviation of the signal. The equivalent average value model is constructed to analyze the influence of the electric vehicle drive system on the zero-sequence circulation. The PID control and fuzzy PID control are effectively combined to establish the electric vehicle drive controller, and then the electric vehicle drive control is completed. The experimental results show that the proposed method can achieve good driving control effect of electric vehicles.

Aiming at the problems of low accuracy and information loss using single sensor for obstacle avoidance in UAVs (Unmanned Aerial Vehicles), a UAV autonomous obstacle avoidance method based on multi-sensor fusion was proposed in this paper. The improved Bayesian fusion algorithm is used to fuse the point cloud acquired by 2D lidar and depth camera to compensate for the areas that the 2D lidar cannot detect. At the same time, an octree map is generated based on the fused point cloud, and the UAV is replanned in real-time according to the updated map information to achieve autonomous obstacle avoidance in unknown environments. The experimental results show that the proposed method not only improves the accuracy of UAV perception of the surrounding environment, with the root mean square error of the fused point cloud is less than 0. 06 m, but also has good obstacle avoidance performance, with the distance between the UAV and obstacles is greater than 0. 5 m, ensuring its safe flight in unknown environments.

The errors in manufacturing, installation, mechanical transmission and temperature change are relatively large, resulting in the offset of the metamorphic center of the compensation robot and the loss of the desired position and posture during the task execution. In order to improve the working efficiency of the welding robot, a compensation control method for the metamorphic center of the welding robot using multi-sensor laser vision is proposed. The welding robot metamorphic center model is established. The laser sensor is used to scan the geometric characteristics of the actual trajectory and posture of the welding robot, and the laser vision image is obtained to calibrate the pipeline weld.The PLC controller is used to adjust the trajectory and posture of the welding robot, eliminate the interference of inertial force on the displacement of the metamorphic center, and realize the compensation control of the welding robot metamorphic center. The experimental results show that the error of the compensation track of the metamorphic center on the X-Y coordinate plane, Z-X coordinate plane and Z-Y coordinate plane is about 5 cm from the expected track,and the height coincides with each other; After control, the inertia displacement of the welding robot is within 0. 1 cm,which is close to zero. The inertia force no longer guides the welding robot to move, and it can stably stop at the target position. The metamorphic center control performance is good.