Contents
2024
Volume: 45 Issue 3
44 Article(s)
Export citation format
Research developments in camouflage object detection
ZHANG Dongdong, WANG Chunping, and FU Qiang
Camouflaged object detection is an important and challenging task which aims to accurately detect tar- gets which are“ perfectly ”hidden in the surrounding environment. Currently , camouflaged object detection has attrac- ted widespread attention in the field of computer vision , and scholars have successfully proposedCamouflaged object detection is an important and challenging task which aims to accurately detect tar- gets which are“ perfectly ”hidden in the surrounding environment. Currently , camouflaged object detection has attrac- ted widespread attention in the field of computer vision , and scholars have successfully proposed various types of detec- tion models. However , most of the work is aimed at building efficient detection models , and there is a lack of in-depth analysis and generalization of existing models. Therefore , this paper presents a comprehensive analysis and summary of existing camouflaged object detection models and discusses potential research directions for camouflaged object detec- tion. Firstly , an overall review of the existing models is given in two broad categories , traditional methods and deep learning-based methods , and the principles , advantages and disadvantages of the relevant models are elaborated ; Sec- ondly , common datasets and evaluation metrics in the field of camouflaged object detection are introduced ; Then , ex- isting deep learning-based camouflaged object detection models are reproduced , and the detection results of different types of models on public datasets are compared in both qualitative and quantitative terms ; Finally , the whole paper is summarized and future research directions in the field of camouflaged object detection are prospected..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 1 (2024)
A review of research progress in weakly supervised semantic segmentation of 3D point clouds
WU Jie, ZHANG Ansi, LI Song, ZHANG Bao, and ZHANG Yizong
In the 3D point cloud semantic segmentation task , using a small amount of labeled point cloud data for semantic segmentation can save the cost of human labeling , and has attracted widespread attention from the academic community in recent years. Traditional 3D point cloud semantic segmentation methods mostly use fullIn the 3D point cloud semantic segmentation task , using a small amount of labeled point cloud data for semantic segmentation can save the cost of human labeling , and has attracted widespread attention from the academic community in recent years. Traditional 3D point cloud semantic segmentation methods mostly use fully supervised methods , which often require manpower and time to label a large number of point clouds , while using weakly super- vised methods only requires a small amount of labeling on point clouds to achieve the same purpose as fully supervised methods. This paper reviews and discusses the development of weakly supervised semantic segmentation of 3D point cloud in recent years , and summarizes the related methods of weakly supervised semantic segmentation from different perspectives. Based on these methods , the results are quantitatively analyzed and discussed on four public datasets. Finally , the challenges of weakly supervised semantic segmentation of 3D point cloud are summarized , and the future development direction is prospected..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 14 (2024)
Application progress of laser-induced breakdown spectroscopy in semiconductor material detection
WU Chuanqi, and XIU Junshan
For the study of semiconductor materials , the element composition and content inside the material is an important factor affecting its performance , although the conventional detection method can detect the composition and content information in the sample , but in this process there are high cost , low efficiency , lFor the study of semiconductor materials , the element composition and content inside the material is an important factor affecting its performance , although the conventional detection method can detect the composition and content information in the sample , but in this process there are high cost , low efficiency , long time , cumbersome process and other problems. In today's evolving field of analytical technology , there is an urgent need to find a sensing technology that is portable , novel and compatible with today 's materials. Based on the above needs , laser-induced breakdown spectroscopy ( LIBS) technology has gradually entered the public's field of vision , the author briefly intro- duced LIBS technology , on this basis , he focused on the use of LIBS technology to analyze the surface of semiconduc- tor materials microanalysis research and semiconductor material metal oxide nanofilm research progress , introduced the advantages of LIBS technology in these two aspects of detection and the future development of LIBS technology in these aspects..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 24 (2024)
Development of a non-destructive testing system for the irradiation resistance of semiconductor lasers
WANG Shuangzheng, CAO Junsheng, YU Songqun, and GAO Zhijian
The electrical derivative and low-frequency electrical noise parameters of semiconductor lasers can re- flect the internal defects of devices and are related to the radiation resistance performance of devices. This thesis intro- duces a irradiation resistance performance testing system for semiconductor lasers designedThe electrical derivative and low-frequency electrical noise parameters of semiconductor lasers can re- flect the internal defects of devices and are related to the radiation resistance performance of devices. This thesis intro- duces a irradiation resistance performance testing system for semiconductor lasers designed based on the electrical de- rivative and low-frequency electrical noise techniques , which can measure and extract the electrical conductivity and low-frequency electrical noise parameters of lasers. By evaluating the sensitive parameters of semiconductor lasers be- fore and after exposure to small doses of radiation , the irradiation resistance performance can be evaluated with the ad- vantages of sensitivity and non-destructive..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 30 (2024)
Design of large aperture continuous zoom optical system based on Cassegrain system
LIU Hao, ZHAO Yu, CHEN Yuanyuan, YAN Hongyu... and LI Jinhua|Show fewer author(s)
:The zoom system can search for the target with a large field of view , but the off-axis aberration of the large field of view will seriously affect the off-axis imaging of the card structure. In order to achieve the requirements of continuous zoom with a large field of view and a long focal length at the same time , a:The zoom system can search for the target with a large field of view , but the off-axis aberration of the large field of view will seriously affect the off-axis imaging of the card structure. In order to achieve the requirements of continuous zoom with a large field of view and a long focal length at the same time , a corrective lens is added to the image side of the card structure , and an infrared continuous zoom system with a working field of 0 ° ~ 1. 5 ° and a focal length range of 800 mm ~ 1 600 mm is designed. The system is composed of seglin telescope structure and three-com- ponent mechanical compensation zoom structure system. The working wavelength is 1 064 nm , the maximum aperture is 300 mm ,and maximum distortion at each focal length is less than 0. 5% . At 17 lp/mm , the MTF at each focal length is greater than 0. 5 , which is close to the diffraction limit. The size of the diffuse spots is very small , all of which are in Airy spots. The aberrations in all fields of view are corrected , and all the indexes meet the design requirements..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 35 (2024)
Study on signal extraction of coherent wind lidar
LIU Zhiping, TIAN Wenfei, HU Yuanhang, FENG Zhenzhong, and TONG Wenhao
The signal extraction method of wind lidar affects the accuracy and range of lidar directly. In order to improve the accuracy and rang of wind lidar , A method of gravity extraction based on Gauss fitting is designed. Firstly frequency spectrum is fitted by Gauss function. Then center of gravity frequency is calculatedThe signal extraction method of wind lidar affects the accuracy and range of lidar directly. In order to improve the accuracy and rang of wind lidar , A method of gravity extraction based on Gauss fitting is designed. Firstly frequency spectrum is fitted by Gauss function. Then center of gravity frequency is calculated for the fitted frequency spectrum. Using the fixed frequency sine signal outputted by the signal source , the average measurement error of bary- center method based on Gaussian fitting is 0. 43 MHz , the average measurement error of the maximum method is 1. 54 MHz and the average error of the barycenter method is 1. 18 MHz. The measured echo signals of the wind lidar are tested. The measuring range of the wind lidar is 2 850 meters base on the Gaussian fitting method and the measuring range of the wind lidar is 2 550 meters. The results show that the measuring range of the wind lidar based on the Gaussian fitting method can improve the accuracy of the wind lidar. And increase the measuring range of wind lidar by 12. 5% 。.
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 40 (2024)
A filterless scheme of generating frequency 16-tupling mm Wave based on triple-parallel MZM
PANG Yanting, MA Huifang, YAN Yingce, and LIU Guoqiang
In order to improve the number of mmWave multiples generated and reduce the system complexity ,a fil- terless 16-tupling millimeter-wave generation scheme based on triple-parallel Mach Zehnder modulator( MZM) is pro- posed. This scheme uses two MZMs in parallel to generate the carrier and the 8th order optical sidebandIn order to improve the number of mmWave multiples generated and reduce the system complexity ,a fil- terless 16-tupling millimeter-wave generation scheme based on triple-parallel Mach Zehnder modulator( MZM) is pro- posed. This scheme uses two MZMs in parallel to generate the carrier and the 8th order optical sidebands , suppresses the carrier by adjusting the DC bias voltage of the third MZM , and obtains a 16-tupling millimeter-wave signal by the photodetector( PD) beating frequency. The principle of 16-tupling millimeter -wave generation is analyzed in detail , and the feasibility of the scheme is verified by simulation. The optical sideband suppression ratio ( OSSR) is achieved 29. 7 dB. All at once ,the value of radio frequency stray suppression ratio ( RFSSR) is 24. 6 dB. According to the simu- lation results , the influence of the shift of modulation index , modulator extinction ratio , DC voltage offset of MZM- 3 and phase difference of MZM- 1 and MZM-2 ratio frequency signals on the system is analyzed , and the results show that the system can still obtain good performance within a certain deviation range..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 45 (2024)
Nondestructive testing of microstructure and properties of aluminum alloy based on laser ultrasound
XU Zhi, ZHANG Zewen, LU Yujie, ZHENG Lei... and YIN Anmin|Show fewer author(s)
At present , scanning electron microscopy ( SEM) and transmission electron microscopy ( TEM) are mainly used to detect the precipitated phase and mechanical properties of aluminum alloy , which is time -consuming and laborious and destructive. In this paper , the laser ultrasonic nondestructive testing technique was usAt present , scanning electron microscopy ( SEM) and transmission electron microscopy ( TEM) are mainly used to detect the precipitated phase and mechanical properties of aluminum alloy , which is time -consuming and laborious and destructive. In this paper , the laser ultrasonic nondestructive testing technique was used to detect Al -Cu-Mg aluminum alloy. By means of Variational Modal Decomposition ( VMD) , the noise reduction of the laser ul- trasonic signal in the time domain is processed , and the attenuation coefficient and sound velocity of the laser ultrasonic signal are obtained. The precipitated phase and mechanical properties of Al-Cu-Mg aluminum alloy were obtained by SEM , electron back scattering diffraction ( EBSD) and static tensile test. The laser ultrasonic signal characteristic val- ues were coupled with the precipitated phase and mechanical properties of Al- Cu - Mg aluminum alloy. The results show that the characteristic value of laser ultrasonic signal has a good correlation with the precipitated phase , grain size , yield strength and microhardness of Al-Cu-Mg aluminum alloy. This provides a new method for rapid nonde-structive testing of microstructure and mechanical properties of Al-Cu-Mg aluminum alloy..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 50 (2024)
Based on the improved YOLOv5 lightweight tank target detection algorithm
MEI Likun, CHEN Zhili, and LI Dongqi
The application of deep learning techniques significantly enhances the accuracy and speed of tank target recognition in military operations , reducing misjudgments and omissions , thereby minimizing casualties and war losses. To address the limitations of large , complex , and computationally intensive models in terms The application of deep learning techniques significantly enhances the accuracy and speed of tank target recognition in military operations , reducing misjudgments and omissions , thereby minimizing casualties and war losses. To address the limitations of large , complex , and computationally intensive models in terms of computing resources , storage , and energy consumption , a YOLOv5-based lightweight object detection system was proposed. This approach enriches gradient flow information and further accelerates computations through the C2f module based on attention mechanism. The combination of Lead Head and Aux Head balances positive and negative samples , improving the mod- el's ability to detect obscured small tank targets. Additionally , the utilization of FasterNet as the feature extraction net- work resolves issues related to high parameter quantity and computational demands. Experimental results demonstrate that compared to the original YOLOv5 , the improved model achieves a 1. 2% and 4. 2% increase in Map0. 5 and mAP0. 5 :0. 95 , respectively , while reducing parameters , GFLOPs , and Best. pt by 32. 3% , 27. 59% , and 26. 01% . The improved YOLOv5 model enables fast and accurate tank target recognition , making it more accessible for deploy- ment on mobile and embedded devices due to its lightweight nature..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 59 (2024)
Improved edge detection algorithm and its FPGA implementation
WANG Zishuo, SHAN Yanhu, CHU Chengqun, CHENG Hongtao... and ZHAO Xinglong|Show fewer author(s)
Aiming at the traditional edge detection algorithm implemented on FPGA , there are problems of lack of adaptivity and poor real-time performance , combined with the characteristics of high-speed parallel processing data on FPGA , an improved edge detection algorithm is proposed. The algorithm firstly applies bilateral Aiming at the traditional edge detection algorithm implemented on FPGA , there are problems of lack of adaptivity and poor real-time performance , combined with the characteristics of high-speed parallel processing data on FPGA , an improved edge detection algorithm is proposed. The algorithm firstly applies bilateral filtering to the Canny edge detection algorithm , which retains the gradient information of the image edge while reducing the noise ; secondly , it uses the improved Sobel operator to calculate the gradient magnitude in four directions to improve the localization ac- curacy of the image edges ; finally , it proposes an adaptive threshold selection strategy based on the change of the first -order derivatives of the gradient histogram in order to improve the adaptive capability of the algorithm. The experi- mental results show that the peak signal-to-noise ratio ( PSNR) of the image after the bilateral filtering process is im- proved by 109% compared with the traditional Canny algorithm , and the single edge response and connectivity of the edges are significantly improved , while meeting the real-time requirements. Provide a feasible solution for the applica- tion of Canny algorithm to embedded machine vision..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 74 (2024)
Study on dam leakage AI detection method based on infrared thermal imaging
WANG Jiahao, DING Yong, HUANG Yinghao, WANG Yi, and WU Yulong
In order to solve the problem of dam leakage detection , this paper presents a dam leakage detection method which combines active excitation infrared imaging with depth learning. The infrared image of leakage is pro- duced by computer simulation , and then combined with the infrared image acquired by simulating dam leaIn order to solve the problem of dam leakage detection , this paper presents a dam leakage detection method which combines active excitation infrared imaging with depth learning. The infrared image of leakage is pro- duced by computer simulation , and then combined with the infrared image acquired by simulating dam leakage test , the infrared image data set is generated for the training of depth learning. On the basis of Yolov5 original model , using AF-FPN to replace FPN can improve the ability of identifying the leakage area of dam infrared image , and make an ef- fective trade - off between identifying speed and accuracy. The test results show that the accuracy of the model is 87. 6% , the recall rate is 96. 5% , and the average accuracy ( IoU = 0. 5) is 88. 3% , which indicates that the method proposed in this paper can identify the leakage area of dam infrared image well..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 81 (2024)
teel surface defect detection based on YOLOv7
ZHANG Yateng, and HUANG Jun
A steel surface defect detection algorithm based on improved YOLOv7 is proposed to address the high false detection rate and missed detection rate in steel surface defect detection. In this algorithm , the ConvNeXt-CBS module is introduced to enhance the feature extraction capability of the network. Additionally , the A steel surface defect detection algorithm based on improved YOLOv7 is proposed to address the high false detection rate and missed detection rate in steel surface defect detection. In this algorithm , the ConvNeXt-CBS module is introduced to enhance the feature extraction capability of the network. Additionally , the MPCS module is constructed based on the SimAM attention mechanism to increase the network ’s focus on small defect targets. Finally , the C3 module is introduced in the model to improve network stability. Experimental results show that on the NEU - DET dataset , the detection accuracy of this algorithm reaches 80. 2% , which is 3. 9% higher than the YOLOv7 algo- rithm. Compared to previous steel surface defect detection algorithms , this algorithm achieves higher detection accura- cy and faster detection speed , making it highly suitable for industrial applications..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 87 (2024)
Optical glass curved lens based on YOLOv5s network Defect detection methods
LIU Xiaolei, LIU Fenghui, and CUI Chenjia
In order to improve the accuracy of surface defect detection for curved optical glass lenses , automated surface defect detection technology is studied. By analyzing the imaging principles of different defects on the surface of optical glass lenses , a defect collection device combining two lighting methods is designedIn order to improve the accuracy of surface defect detection for curved optical glass lenses , automated surface defect detection technology is studied. By analyzing the imaging principles of different defects on the surface of optical glass lenses , a defect collection device combining two lighting methods is designed to capture high contrast de- fect images while compensating for the shortcomings of optical lens defect detection in detachment defects;Preprocess and enhance the collected defect images to provide high-quality images for automated defect detection of optical glass lens;Applying deep learning methods to optical lens defect detection , by comparing the performance of different net- work models on the optical lens defect dataset ,Select the YOLOv5s with the best performance to complete the detection of lens defects , with a recall rate and average accuracy of 92% and 95% , respectively. The time to detect a defective lens is 10 ms..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 94 (2024)
Small target detection based on traffic signs
ZENG Tianhao, and CHEN Lin
Due to the low accuracy and false detection of small target detection algorithm in traffic sign detection , a new foreground fusion attention mechanism network called YOLO-Traffic is proposed. First , EIOU loss function is introduced to calculate the width of the predicted frame and the real frame respectively , and thDue to the low accuracy and false detection of small target detection algorithm in traffic sign detection , a new foreground fusion attention mechanism network called YOLO-Traffic is proposed. First , EIOU loss function is introduced to calculate the width of the predicted frame and the real frame respectively , and the dilated convolution is used to solve the problems existing in the original CIOU model. Secondly , the foreground attention mechanism F-ECA was added to fully extract the foreground information and suppress background noise. Finally , Kmeans+ + algorithm is used to replace the anchor frame obtained by Kmeans clustering to reallocate the corresponding feature layer and further improve the feature extraction ability. The experiment on TT100K traffic sign data set produced by Tsinghua University shows that compared with the original YOLOv5 network and the accuracy is increased by 2. 91% , the recall rate is in- creased by 2. 1% , the detection speed is 44 frames per second , and the final accuracy reaches 96. 89% . Hence , the proposed YOLO-Traffic network promotes the accuracy of traffic sign detection and model performance..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 100 (2024)
Research on magnetic field measurement method based on laser self-mixing interference
ZHAO Yan, LI Kangda, LIN Maohua, SUN Wenqiang, and ZHANG Zhengyang
In view of the problems that optical fiber MHD magnetic field sensor is susceptible to environmental in- terference and the MHD film production process is complex , laser self-mixing interference has the characteristics of strong anti-interference ability and good stability , a magnetic field measurement method based oIn view of the problems that optical fiber MHD magnetic field sensor is susceptible to environmental in- terference and the MHD film production process is complex , laser self-mixing interference has the characteristics of strong anti-interference ability and good stability , a magnetic field measurement method based on spatial light horizon- tal polarization laser self-mixing interference is proposed. In a certain range , the magnetic field is proportional to the change of the signal amplitude , and the magnetic field is detected by using the mean ratio of the signal amplitude be- fore and after the magnetic field is applied. The experimental results show that the amplitude of signal attenuates obvi- ously in MHD with 1% concentration , and the film thickness is 0. 5 mm , which reduces the difficulty of making the film. In the experimental results , the mean ratio of 5. 02 mt to 19. 112 mt magnetic field intensity was linearly fitted , and the relative error was at least 0. 17% ..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 106 (2024)
Lightweight optical remote sensing image road extraction based on L-DeepLabv3+
XIE Guobo, HE Lin, LIN Zhiyi, ZHANG Wenliang, and CHEN Yi
To address the problems of large number of model parameters and poor detail extraction in DeepLabv3+ for optical remote sensing image road extraction task , a light-weight road extraction model L-DeepLabv3+ is proposed to improve DeepLabv3+. Firstly , the number of model parameters is reduced by replacing the backbone To address the problems of large number of model parameters and poor detail extraction in DeepLabv3+ for optical remote sensing image road extraction task , a light-weight road extraction model L-DeepLabv3+ is proposed to improve DeepLabv3+. Firstly , the number of model parameters is reduced by replacing the backbone network with MobileNetv2 ; secondly , an improved void space convolutional pooling pyramid module is designed in the coding layer. This module enhances the model feature expression capability by embedding a channel space parallel attention module and YOLOF module , and replaces the normal convolution with deep separable convolution to further reduce the number of model parameters ; Finally , Dice_loss and Focal _loss are combined as loss functions to solve the positive and nega- tive sample imbalance problem. The experimental results show that L-DeepLabv3+ achieves 68. 40% intersection ratio and 82. 67% pixel accuracy for road extraction on DeepGlobe Road dataset , and the number of model parameters is on- ly 5. 63 MB , and the FPS reaches 72. 3 , which is a significant improvement compared with other models , and achieves a better balance between model accuracy and light weight..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 111 (2024)
Multiscale fusion single image superresolution reconstruction based on attention mechanism
SHENG Yue, XIN Yuelan, WANG Qingqing, and XIE Qiqi
Aiming at the issues of inadequate feature extraction and insufficient ability to reconstruct high-frequen- cy details in the information recovery process of image super-resolution reconstruction algorithm , a multi-scale fused image super-resolution reconstruction algorithm ( SRGAN-MCA) based on the attention mechanisAiming at the issues of inadequate feature extraction and insufficient ability to reconstruct high-frequen- cy details in the information recovery process of image super-resolution reconstruction algorithm , a multi-scale fused image super-resolution reconstruction algorithm ( SRGAN-MCA) based on the attention mechanism is proposed on the basis of SRGAN. First , a multi-scale dense residual attention module based on coordinate attention mechanism is con- structed to extract feature information at different scales to solve the problem of inadequate feature extraction in the process of nonlinear mapping of image super-resolution reconstruction ; second , the Lipschitz constant of the discrimi- nator is constrained by embedding spectral normalization in the network discriminator to enhance the stability of net- work training; finally , the Charbonnier loss function to SRGAN-MCA for training optimization to achieve higher quality reconstruction. The experimental results on Set5 , Set14 , and BSD100 datasets show that the peak signal-to-noise rati- o ( PSNR) is improved by 0. 35 dB and 0. 47 dB on average , and the structural similarity ( SSIM) is improved by 0. 006 and 0. 016 on average for the 2 and 4 magnification reconstructed images compared with SRGAN..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 118 (2024)
MRI subtle feature extraction method based on fusion Canny-SIFT algorithm
LIU Haoyu, LIU Xiaobao, YAO Tingqiang, and SHEN Jihong
Aiming at the problem that the subtle features in magnetic resonance images ( MRI) are difficult to ex- tract and easy to be missing , a method for extracting subtle features of magnetic resonance images based on the fusion of Canny-SIFT algorithm is proposed. The algorithm first solves the problem of unclear texture aAiming at the problem that the subtle features in magnetic resonance images ( MRI) are difficult to ex- tract and easy to be missing , a method for extracting subtle features of magnetic resonance images based on the fusion of Canny-SIFT algorithm is proposed. The algorithm first solves the problem of unclear texture and subtle feature infor- mation in the image due to the uneven gray level of the image and the complex noise signal. It uses automatic Gamma transformation to increase the image contrast. Separately deal with the noise of the observation area; for the incomplete extraction of subtle features in the key observation area , the importance is divided according to the actual diagnosis re- quirements , and the location , area and other information of the important area are obtained through feature matching and topological relationship reasoning , and the self-adaptation is completed at the same time Select the corresponding Sobel operator; finally , the output image is obtained by thresholding segmentation and binarization ; Experiments show that the proposed method has significantly improved the edge detection accuracy compared with the existing Canny method , the structural similarity is increased by 38% , and the mean squared error is reduced by 31. 4% , and the per- formance is the best compared with other mentioned algorithms..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 126 (2024)
Improved segmentation of unstructured road image instances in SOLOv2
SONG Liang, GU Yuhai, and HUANG Jiawei
In view of the overlapping multiple targets in unstructured road images and the large difference in scale , it is easy to miss or misdetect and poor segmentation accuracy. An improved SOLOv2 instance segmentation algorithm is proposed. Firstly , a bottom-up enhancement path is added to the feature pyramid structure to In view of the overlapping multiple targets in unstructured road images and the large difference in scale , it is easy to miss or misdetect and poor segmentation accuracy. An improved SOLOv2 instance segmentation algorithm is proposed. Firstly , a bottom-up enhancement path is added to the feature pyramid structure to reduce the loss of fea- ture transfer process , and secondly , dual attention is used to guide feature selection , adaptive selection of important features , suppression of redundant information , improvement of the extraction ability of detailed features , and enhance- ment of feature representation of category branches and mask branches , so as to improve the accuracy of mask predic- tion. In addition , the unstructured road image dataset is preprocessed to improve the generalization ability of the mod- el. The experimental results show that the proposed method is more accurate in controlling the instance boundary , and the average accuracy of SOLOv2 and Mask-RCNN is increased by 2. 0% and 2. 2% , respectively , and the detection frame rate is increased to 6. 1 frames/s , which has good segmentation performance in different environments..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 133 (2024)
GPU parallel acceleration of two-channel alternating minimization algorithm
HAN Xue, LIU Jinlong, LI Songheng, YANG Huizhen... and LI Ziwei|Show fewer author(s)
Due to atmospheric turbulence and system noise , the images of astronomical or space objects are blurred and degraded. The dual channel alternating minimization algorithm is one of the effective methods for restoring images degraded by turbulent and noise. However , this algorithm is relatively complex and requires repDue to atmospheric turbulence and system noise , the images of astronomical or space objects are blurred and degraded. The dual channel alternating minimization algorithm is one of the effective methods for restoring images degraded by turbulent and noise. However , this algorithm is relatively complex and requires repeated iterative operations , resulting in a longer processing time. In order to improve the algorithm running speed , the graphics proces- sor ( GPU) acceleration technology based on the algorithm structure features is applied to the dual channel alternating minimization algorithm , with a focus on optimizing the iterative process of alternating minimization. The experimental results show that under the condition of different atmospheric turbulence and Signal-to -noise ratio 20 dB ,compared with the algorithm directly using the central processing unit ( CPU) , GPU parallel acceleration for the dual channel al- ternating minimization algorithm can achieve the " U-step" operation rate of image restoration increased by more than 80% , and the " H-step" operation rate of point spread function solution increased by more than 60% , and the recon- structed images are close to the diffraction limit. The combination of parallel acceleration technology and existing algo- rithms can effectively improve the running speed , providing a certain reference for the restoration of degraded images caused by turbulence and noise..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 140 (2024)
Research on laser video image retrieval based on mutual information mean square error extraction of key frames
HU Xiu, and WANG Shu’ai
In order to ensure that there are no repetitive redundant images in laser video image retrieval results , a laser video image retrieval method based on mutual information mean square error extraction of key frames is studied.In this method , a key frame extraction method based on the mutual information mean square erroIn order to ensure that there are no repetitive redundant images in laser video image retrieval results , a laser video image retrieval method based on mutual information mean square error extraction of key frames is studied.In this method , a key frame extraction method based on the mutual information mean square error of laser video image color is adopted to maximize the mutual information mean square error of laser video image to set the criteria for the clustering center of laser video image key frame , so as to extract non-repetitive video image key frames by clustering. Through the laser video image retrieval method based on key frame , the extracted key frame is taken as the core judg- ment content of laser video image retrieval , and the laser video image with significant similarity to the required image key frame is extracted to complete the laser video image retrieval. The experimental results show that the redundancy of key frames extracted by this method is only 0. 01 , and the test value of MAP index of laser video image retrieval results is as high as 0. 98. There is no repetitive redundant image in the retrieval results..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 145 (2024)
Video image geometric distortion correction method based on optical fiber sensor
ZHANG Xiguang, ZHANG Longfei, and FAN Yinting
During the propagation process of video images , information distortion may occur , resulting in geomet- ric distortion of the image. Therefore , a video image geometric distortion correction method based on fiber optic sensors is proposed. Analyze the distribution of pixel nodes in the photosensitive area of the videoDuring the propagation process of video images , information distortion may occur , resulting in geomet- ric distortion of the image. Therefore , a video image geometric distortion correction method based on fiber optic sensors is proposed. Analyze the distribution of pixel nodes in the photosensitive area of the video image , extract sub pixel dis- parity information , and obtain distorted video images through an array CCD image sensor; Using a distortion curve ap- proximation model to calculate the distortion rate of video images , determine the coordinates of the distorted images , and combine bilinear grayscale interpolation algorithm to derive correction parameters to achieve geometric distortion correction of video images. The experimental results show that the proposed method can effectively correct geometric distortions of different types of video images. The similarity between the corrected image and the original image exceeds 0. 97 , and the degree of image edge blur is reduced , improving the visual observation effect of the image..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 150 (2024)
Research on laser image edge detection based on strong noise
YANG Tianmin, and WANG Xiaoyan
Edge detection is an important step in laser image processing. In order to obtain complete laser image edges , a laser image edge detection method based on strong noise is designed for the defects such as large error in laser image edge detection and easy to be disturbed by noise. First , the laser image is collected ,Edge detection is an important step in laser image processing. In order to obtain complete laser image edges , a laser image edge detection method based on strong noise is designed for the defects such as large error in laser image edge detection and easy to be disturbed by noise. First , the laser image is collected , and the histogram equaliza- tion technology is used to enhance the original laser image to improve the quality of the laser image. Then , wavelet transform is used to remove the noise of the laser image , and the Canny operator is used to detect the edge of the laser image. Finally , the test results show that compared with the current laser image edge detection method , this method can detect more details of the laser image edge , The average value of laser image edge detection accuracy is more than 92% , and the laser image edge detection time is controlled within 60 ms. Laser image provides a high-precision detec- tion technology..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 156 (2024)
Research on registration of multi-view point cloud based on depth sensor
LIU Yaowen, BI Yuanwei, ZHANG Lujiang, and HUANG Yansen
To address the issues of mismatched point pairs and cumulative errors encountered during the multi - view point registration of large - scale symmetrical objects , a multi-view point cloud registration algorithm based on depth sensor is proposed. Firstly , the proposed approach leverages depth sensors to capture multipTo address the issues of mismatched point pairs and cumulative errors encountered during the multi - view point registration of large - scale symmetrical objects , a multi-view point cloud registration algorithm based on depth sensor is proposed. Firstly , the proposed approach leverages depth sensors to capture multiple point clouds of the target object from various viewpoints , which are then subjected to a series of preprocessing steps. To achieve coarse registration , the Super 4-points congruent sets ( Super4PCS) is employed specifically for adjacent point clouds on one side of the object. Subsequently , an enhanced point-to-plane ICP algorithm is utilized to refine the registration by e- liminating erroneous point pairs. The resulting refined point clouds from the left and right sides are seamlessly com- bined , thereby generating a comprehensive 3D point cloud model. Furthermore , to mitigate the issue of cumulative er- rors arising from the multi-view registration process , a global optimization technique is introduced. Experimental eval- uations demonstrate the effectiveness and accuracy of the proposed method in achieving precise multi-view point cloud registration and generating highly accurate 3D point cloud models..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 161 (2024)
COVID-19 lesion segmentation network based on multi-scale feature fusion attention
LIN Jieqin, and HUANG Xin
Because of COVID- 19's highly infectious , early diagnosis and treatment are the key factors to reduce the losses caused by the epidemic. In order to assist doctors in the diagnosis of COVID- 19 and efficiently segment CO- VID- 19 lesions from lung CT slices , an improved encoder-decoder deep neural network based oBecause of COVID- 19's highly infectious , early diagnosis and treatment are the key factors to reduce the losses caused by the epidemic. In order to assist doctors in the diagnosis of COVID- 19 and efficiently segment CO- VID- 19 lesions from lung CT slices , an improved encoder-decoder deep neural network based on the U-Net with ex- cellent image segmentation effect , Multi-scale Attention Network ( MSANet ) is proposed. By using a global pooling layer and setting a sampling rate for void convolution , the network receptive field is increased , and multiscale informa- tion is captured to achieve effective segmentation of large objects. MSANet uses channel attention and spatial attention to model in the spatial dimension to effectively extract deep image features. The test results show that compared with U -Net network , the improved algorithm improves the mean intersection over union of segmentation by 1. 46% , improves the mean pixel accuracy of category by 0. 8% , and improves the accuracy by 1. 17% ..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 168 (2024)
Research on dual branching decoding lightweight segmentation networks based on unmanned ships
LIU Dan, and ZHANG Jianjie
In order to ensure smooth navigation of unmanned surface vessels ( USVs) for water missions , accurate extraction of river information is required , so a semantic segmentation network model of river is investigated. To ad- dress the problem of inter-class inconsistency and intra-class inconsistency in river image segmeIn order to ensure smooth navigation of unmanned surface vessels ( USVs) for water missions , accurate extraction of river information is required , so a semantic segmentation network model of river is investigated. To ad- dress the problem of inter-class inconsistency and intra-class inconsistency in river image segmentation , a segmenta- tion network DBDL- Net is proposed in the paper , in which a double -branch decoding structure and a double loss function are designed to capture semantic and spatial information respectively; a lightweight module with multi-scale residuals is also designed in the coding part to reduce parameters on the one hand and capture feature information at different scales on the other. Finally , the model is ablated and compared with experiments on the USVInland dataset. The experimental results show that the accuracy and the mIoU of DBDL-Net finally reach 93. 619% and 87. 682% , and DBDL-Net also has better overall performance compared with other advanced segmentation networks..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 175 (2024)
Study on single mode coupling efficiency of fiber coupled optical systems
YU Xishuang
Single-mode coupling is a key technology in the fiber -coupled optical system. At present , there are still some problems in the process of single-mode coupling , which directly affect the results of single-mode coupling. In order to obtain better single-mode coupling results , the single-mode coupling efficiency of thSingle-mode coupling is a key technology in the fiber -coupled optical system. At present , there are still some problems in the process of single-mode coupling , which directly affect the results of single-mode coupling. In order to obtain better single-mode coupling results , the single-mode coupling efficiency of the fiber-coupled optical system is analyzed and studied in this paper. Firstly , a fiber coupled optical system is designed according to the princi- ple of single mode coupling , and the transmitter and receiver are designed. Then the single mode coupling model of the fiber coupled optical system is constructed. The test results show that the single -mode coupling model of the design system has obvious advantages , and the coupling efficiency has been significantly improved , which provides a valuable reference for the related fiber coupled optical system..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 182 (2024)
Research on phase carrier demodulation method of aperiodic low frequency dynamic signal of optical fiber sensor
YE Xiaoran, CHEN Gaixia, and DONG Huimin
In order to improve the phase carrier demodulation effect of aperiodic low frequency dynamic signal of optical fiber sensor , the phase carrier demodulation method of aperiodic low frequency dynamic signal of optical fiber sensor is proposed. Adaptive beamforming algorithm is used to enhance the details of aperiodic loIn order to improve the phase carrier demodulation effect of aperiodic low frequency dynamic signal of optical fiber sensor , the phase carrier demodulation method of aperiodic low frequency dynamic signal of optical fiber sensor is proposed. Adaptive beamforming algorithm is used to enhance the details of aperiodic low-frequency dynamic signals ; The signal features are extracted by singular value decomposition algorithm; The extracted features are input into the CD3S signal demodulation model based on the double extended Kalman filter , and the noise is eliminated through the Kalman filter structure ; Through joint estimation , the spread spectrum code synchronization of the dynamic signal is realized , and the phase carrier demodulation of the aperiodic low-frequency dynamic signal of the optical fiber sensor is completed. The experimental results show that the demodulation error rate of the proposed method is below 1. 0% with or without noise interference , and the displacement error after demodulation is as low as 12 pm , which im- proves the demodulation effect and anti-noise ability..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 187 (2024)
Laboratory infrared monitoring system based on conditionally generated countermeasure network
WANG Hai, and LIANG Jianfeng
As a scientific research base , the laboratory stores a lot of valuable data and dangerous goods , so it is necessary to carry out safety monitoring. In this context , a laboratory infrared monitoring system based on conditionally generated countermeasures network is designed. The system framework is designed with the As a scientific research base , the laboratory stores a lot of valuable data and dangerous goods , so it is necessary to carry out safety monitoring. In this context , a laboratory infrared monitoring system based on conditionally generated countermeasures network is designed. The system framework is designed with the help of B/S three-layer ar- chitecture , which is divided into acquisition layer , network layer and application layer. MySQL is used as the support to design the database. With STC89C52 single chip microcomputer as the core , it controls the peripheral hardware and builds the system hardware. Based on ZigBee technology , the wireless transmission network of the system is established for remote transmission. The function unit of the monitoring system is designed. The infrared image of the laboratory is collected by the acquisition unit , and the pretreatment unit is used to carry out noise removal , brightness adjustment , and image segmentation pretreatment for the infrared image of the laboratory. The anomaly recognition unit is designed based on the condition generation countermeasure network ( cGAN) to determine whether there is any anomaly in the monitoring image and give early warning based on it. The results show that the monitoring results of the system are con-.
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 193 (2024)
Research on optical communication signal recognition based on blockchain technology
YANG Jianli, SUN Xia, and ZHAO Pan
As an emerging communication method in the field of wireless communication , optical communication has broad research prospects. The study of its system signal perception method is a key issue in the field of optical communication. At this stage , the signal perception method of optical communication has the problem ofAs an emerging communication method in the field of wireless communication , optical communication has broad research prospects. The study of its system signal perception method is a key issue in the field of optical communication. At this stage , the signal perception method of optical communication has the problem of poor recogni- tion effect and great interference with noise. In order to solve the problems in the solution , the light communication system signal intelligent perception method based on blockchain technology is proposed. First of all , use the Fourier transformation of the Four Dege to collect optical communication signals with a combination of reconstruction methods. Intelligent perception. The experimental results show that the method of signaling the optical communication system mentioned by the method is higher , the noise is smaller , and it has better practical application value..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 199 (2024)
Research on weak signal detection in optical communication based on big data mining
DONG Niya, and LIN Yi
Design an optical communication weak signal detection method based on big data mining with the aim of efficiently and accurately detecting weak signals submerged in noise in optical communication. The noise component in the optical communication signal is completely removed based on improved EMD and singular value decoDesign an optical communication weak signal detection method based on big data mining with the aim of efficiently and accurately detecting weak signals submerged in noise in optical communication. The noise component in the optical communication signal is completely removed based on improved EMD and singular value decomposition. Af- ter the parameters of the support vector machine are optimized by the gray wolf algorithm , the signal classification hy- perplane is constructed from the support vector machine model , and the weak signals in the samples are classified and detected. The experimental results show that after denoising the optical communication signal using the proposed meth- od , the signal-to-noise ratio of the signal decreases , with a maximum value of only 0. 01 dB. The fluctuation ampli- tude of the weak signal detected by the proposed method highly matches the actual amplitude of the weak signal , with an error of no more than 1% , 100% detect samples of weak optical communication signals submerged in noise..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 204 (2024)
Edge computing based evaluation method of optical communication network security risk level
DU Yuhong, and HOU Shouming
For the purpose of improving the security of optical communication network , a method for evaluating the security risk level of optical communication network based on edge computing is proposed. Firstly , the security risk as-sessment index of optical communication network is constructed , and the index data of each teFor the purpose of improving the security of optical communication network , a method for evaluating the security risk level of optical communication network based on edge computing is proposed. Firstly , the security risk as-sessment index of optical communication network is constructed , and the index data of each terminal is calculated using each edge computing node. Based on the calculation results of various evaluation index data , a fuzzy comprehensive e- valuation model is constructed within the cloud center computing layer. Based on the membership degree in fuzzy mathematics theory , multi-level fuzzy evaluation of optical communication network security risks is achieved. The ex- perimental results show that the highest accuracy of the optical communication network security risk level evaluation u- sing this method is 97% , and the lowest error rate is 3% . This method can effectively implement the evaluation process , and network optimization based on the evaluation results can improve the security performance of the optical communication network..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 209 (2024)
Normal vector extraction of laser point cloud boundary based on big data mining
WANG Liyun, CHU Hanbing, QIN Lijuan, and ZHANG Xianjing
In the process of laser point cloud boundary normal vector extraction , the point cloud boundary data is difficult to accurately register , resulting in poor extraction accuracy and efficiency. In order to analyze the laser scan- ning information more accurately , a method of extracting the normal vector of laser pointIn the process of laser point cloud boundary normal vector extraction , the point cloud boundary data is difficult to accurately register , resulting in poor extraction accuracy and efficiency. In order to analyze the laser scan- ning information more accurately , a method of extracting the normal vector of laser point cloud boundary based on big data mining is proposed. This method first calibrates the laser point cloud through the camera to obtain the laser point cloud image , and then de-noises the acquired image through the principal component analysis , surface fitting and fil- tering algorithm. Finally , the image is processed by graying and Gaussian filtering. At the same time , the gray center in the normal direction of the image center point is extracted by combining the Otsu threshold method , so as to achieve the extraction of the normal vector of the laser point cloud boundary. The experimental results show that the included angle feature of normal vector extracted by the proposed method is basically consistent with the included angle feature in the ideal state , and the extraction efficiency is high and the iteration error is small..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 214 (2024)
Phishing detection in laser communication based on multi-source heterogeneous data fusion analysis technology
XU Huanxiao, CHEN Hongyun, WU Fang, and QIAN Lanmei
phishing brings great danger to the application of laser communication network. In order to ensure the security of laser communication network application , a laser communication phishing detection method based on multi- source heterogeneous data fusion analysis technology is proposed. First , collect the characteristiphishing brings great danger to the application of laser communication network. In order to ensure the security of laser communication network application , a laser communication phishing detection method based on multi- source heterogeneous data fusion analysis technology is proposed. First , collect the characteristics of laser communica- tion phishing , use D-S evidence theory to fuse the characteristics of laser communication phishing data from different sources and structures , then use support vector machine to detect and design the detection model of laser communica- tion phishing according to the special number , and finally carry out a simulation experiment. The results show that this method can effectively extract and accurately fuse the phishing feature data , Laser communication phishing detection The accuracy rate of laser communication phishing detection exceeds 95% , reducing the security risk of laser communi- cation network..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 219 (2024)
Effect of CMP process parameters on the substrate transfer of AlGaInP-based LEDs
WANG Jiawei, XU Yingchao, YANG Kai, LU Chendong... and LU Yi|Show fewer author(s)
In this paper , the GaP/Al2 O3/SiO2 guided light-emitting structure is used as the bonding layer of the chip , and the chemical mechanical polishing process is introduced to reduce the epitaxial voids in the transfer process of AlGaInP-based Mini-LED substrates to improve the chip preparation process yields. Using mateIn this paper , the GaP/Al2 O3/SiO2 guided light-emitting structure is used as the bonding layer of the chip , and the chemical mechanical polishing process is introduced to reduce the epitaxial voids in the transfer process of AlGaInP-based Mini-LED substrates to improve the chip preparation process yields. Using material removal rate and surface roughness as technical evaluation indexes , L9(34 ) orthogonal experiments were conducted based on the re- sults of single-factor experiments on polishing pressure , polishing head speed , polishing plate speed , and polishing fluid flow rate. The experimental results show that the material removal rate is 83. 12 nm/min and the surface rough- ness is as low as 0. 477 nm under the conditions of polishing head speed 75 rpm , polishing plate speed 80 rpm , polis- hing pressure 8 kPa and polishing fluid flow rate 100 mL/min. The optimized process conditions can obtain high quali- ty GaAs bonding surfaces , effectively reduce epitaxial voids , and improve preparation yields..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 224 (2024)
A novel safety and reliable platform door control system based on LiDAR detection for high-speed railway
WANG Haodong, WANG Zhifei, LI Fan, and WEI Yaonan
In response to the issue , the existing platform door control systems cannot identify multiple train door locations , particularly for railway train sets without Automatic Train Operation ( ATO) . Different train models operate mixedly at the same platform , so this paper proposes a railway platform door control systemIn response to the issue , the existing platform door control systems cannot identify multiple train door locations , particularly for railway train sets without Automatic Train Operation ( ATO) . Different train models operate mixedly at the same platform , so this paper proposes a railway platform door control system based on LiDAR technolo- gy. Firstly , the arrangement information of door locations in a multi-model railway environment is analyzed. Subse- quently , the train sets' corresponding position information and the stable stopping information are collected using Li- DAR technology , followed by the design of a LiDAR position tracking control unit. The platform door control system is then designed and eventually verified through experiments. The experimental results demonstrate that the system is safe and reliable. It can accurately identify the door positions of different high-speed train units and determine when the train comes to a complete stop. Its accuracy exceeds 98. 4% , providing both theoretical and practical foundations for the promotion and application of platform doors in subsequent railway stations..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 230 (2024)
Research on packing plate size detection system based on line structured light
TENG Chuanchao, and PANG Mao
In order to detect the peak height and average wave spacing of the packing board , a size measurement system based on line structured light is proposed. Firstly , the composition of the system and the measurement scheme based on the laser triangle method were introduced , calibrated the measurement system , obtained thIn order to detect the peak height and average wave spacing of the packing board , a size measurement system based on line structured light is proposed. Firstly , the composition of the system and the measurement scheme based on the laser triangle method were introduced , calibrated the measurement system , obtained the coordinate con- version relationship between the camera imaging model and the measured object. Then , extracted the center of the line laser stripe by the improved gray-gravity method. Finally , the collected point cloud data was sliced along the y-axis to fit points on multiple sections to complete peak height and average wave spacing measurement. The experimental re- sults show that the improved light bar center extraction method has higher accuracy , the measurement error of peak height can reach ± 0. 04 mm , and the measurement error of average wave spacing can reach ±0. 08 mm , which meets the actual production needs..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 237 (2024)
Calibration method of optical axis perpendicularity error of laser probe in nc machining
LIU Jiantong, HUANG Jingde, and FENG Yue
In NC machining , because the verticality error of the optical axis of the laser probe cannot be accurately obtained , in order to solve this problem , a calibration method for the optical axis perpendicularity error of the laser probe in NC machining is proposed. By building a mathematical model , the reference plane In NC machining , because the verticality error of the optical axis of the laser probe cannot be accurately obtained , in order to solve this problem , a calibration method for the optical axis perpendicularity error of the laser probe in NC machining is proposed. By building a mathematical model , the reference plane of the optical axis of the laser probe is determined , and the ideal perpendicularity of the optical axis of the laser probe is calculated. The binary denoising operation is performed on the optical axis image of the laser probe , and expand its boundary. The perpendic- ularity characteristic information of the optical axis of the laser probe is extracted by Harris operator , and the perpen- dicularity between the optical axis and the object surface is adjusted to achieve the calibration of the perpendicularity error of the optical axis of the laser probe. The experimental results show that the proposed method has high accuracy in calibrating the optical axis perpendicularity error of the laser probe , and the error value is kept at - 0. 5 ~ 0. 5 mm , indicating that the proposed method has good performance..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 243 (2024)
Method of laser weld penetration monitoring based on plasma characteristic information
ZHANG Jiachen
Although laser welding has high efficiency , it is difficult to operate. The welded workpiece is prone to defects , incomplete penetration or excessive penetration , leading to unqualified welding quality. Facing this situation , in order to realize the effective inspection of the welding workpiece , it is very necessaAlthough laser welding has high efficiency , it is difficult to operate. The welded workpiece is prone to defects , incomplete penetration or excessive penetration , leading to unqualified welding quality. Facing this situation , in order to realize the effective inspection of the welding workpiece , it is very necessary to implement the laser weld penetration monitoring. In the research , the height camera is used to collect the plasma topography image and carry out four pre-processing to remove the interference information in the image. Four kinds of feature information of plasma to- pography image are extracted , including centroid width , centroid height , centroid swing angle and plasma area. Based on BP neural network , the mapping relationship model between the ionomer feature information and the laser weld pen- etration type is constructed. With the feature information as the input , the penetration probability value is output. The type whose probability value is closer to 1 is the monitored penetration result..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 248 (2024)
Study on gas concentration detection based on near infrared spectroscopy
LIANG Liang, DU Yuxin, and YANG Zijian
The gas concentration is of great significance in the analysis of various fields. Because in the process of gas concentration detection , the impact of spectral dimensions leads to large errors in the detection results. In order to reduce the adverse effects in the measurement process , a gas concentration detection reThe gas concentration is of great significance in the analysis of various fields. Because in the process of gas concentration detection , the impact of spectral dimensions leads to large errors in the detection results. In order to reduce the adverse effects in the measurement process , a gas concentration detection research based on near-infrared spectroscopy technology is proposed. . The near-infrared spectral baseline was corrected by castration - standard nor- mal transformation. Combine the generalized S transform and singular value to decompose and denoise the near-infra- red spectrum to improve the spectral quality. Based on principal component analysis ( PCA) , the partial least squares ( PLS) dimensionality reduction method is proposed for the dimensionality reduction of near - infrared spectroscopy. Based on Lambert Beer's law , Lorenz linear fitting of near-infrared spectral absorption line is introduced , and gradient descent method is used to directly fit the pre-processed near-infrared spectral absorption signal , and the final gas con- centration detection result is calculated. The experimental results show that the detection results of the proposed meth- od are basically consistent with the actual gas concentrations when detecting the concentrations of toluene , propane and propylene , effectively reducing the residual sum of squares and root mean square error , and the detection time is less than 2. 3 s..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 253 (2024)
Influence of residual line birefringence on stability of operating point of fiber optics current transformer
HUANG Junchang, GUO Xianshan, WANG Qing, LEI Min... and WU Chongqing|Show fewer author(s)
The residual line birefringence in the elliptic maintaining fiber is easily affected by the environment , so the performance of the Fiber Optics Current Transformer ( FOCT) is easily affected by the environment. By using qua- ternion method , the problem of operating point instability of FOCT due to residual line birefThe residual line birefringence in the elliptic maintaining fiber is easily affected by the environment , so the performance of the Fiber Optics Current Transformer ( FOCT) is easily affected by the environment. By using qua- ternion method , the problem of operating point instability of FOCT due to residual line birefringence is analyzed theo- retically. The results show that the ratio of circular birefringence to residual line birefringence RCL ( circular/line ratio) directly affects the stability of the operating point , and the general trend is that the operating point becomes more stable with RCL increasing. When RCL <2 , the operating point is extremely unstable ; on the other hand , when RCL >4 it be- comes relatively stable. Therefore , the parameter RCL is an important specification to measure the elliptic maintaining fiber , and should be measured before making the FOCT to ensure the performance of FOCT..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 258 (2024)
Automatic detection method of sports injury tissue based on laser ultrasound technology
XIONG Yi, and LIU Hao
At present , there are problems with large errors in the detection of sports injury tissues. In order to im- prove the detection effect of sports injury tissues , an automatic detection method for sports injury tissues based on laser ultrasound technology is proposed. First , analyze the research status of sports injurAt present , there are problems with large errors in the detection of sports injury tissues. In order to im- prove the detection effect of sports injury tissues , an automatic detection method for sports injury tissues based on laser ultrasound technology is proposed. First , analyze the research status of sports injury tissue detection , find out the rea- son why the current sports injury tissue detection effect is poor , then collect the laser ultrasonic image of sports injury tissue , confirm the target position by principal component analysis , remove the noise of ultrasonic image by Bilateral filter method , finally , carry out automatic detection of sports injury tissue , and carry out the simulation experiment of sports injury tissue detection. The results show that , The method proposed in this article has a high signal-to-noise ra- tio for motion damaged tissue images , which can eliminate noise. The average accuracy of motion damaged tissue area detection is 98. 63% , and the average mean square error of motion damaged tissue detection is about 0. 044..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 265 (2024)
Optimization method of waste glass target sorting based on laser scanning
YANG Bo, XU Zhenming, and LU Yingqi
Waste glass sorting has the problem of poor sorting effect and long time. Therefore , an optimization method for target sorting of waste glass based on laser scanning is proposed. The glass point cloud data is obtained by laser scanning , the original point cloud data is processed into slice data and compressed , and tWaste glass sorting has the problem of poor sorting effect and long time. Therefore , an optimization method for target sorting of waste glass based on laser scanning is proposed. The glass point cloud data is obtained by laser scanning , the original point cloud data is processed into slice data and compressed , and the point cloud data error is controlled by the least square method , so as to obtain the target data of the area to be sorted. The image threshold segmentation method is used to complete the foreground and background segmentation of the regional image by measur- ing the gap between the targets. The quadtree method is used to decompose the control points of the foreground image containing the targets and reconstruct the background. The waste glass features are extracted by smoothing the back- ground image with Gaussian filter. Finally , the template matching method is used to find the abandoned glass targets , so as to achieve the sorting work. The experimental results show that the waste glass sorted by the proposed method is consistent with the actual results , and the sorting time is less than 1. 5 s , which verifies that the method has good sor- ting effect and higher application value..
Laser Journal
- Publication Date: Oct. 15, 2024
- Vol. 45, Issue 3, 269 (2024)
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20