Journals >Optics and Precision Engineering
Contents
2024
Volume: 32 Issue 24
11 Article(s)
Export citation format
Research on high sensitivity methane detection technology in coal mines based on frequency modulation
Xiangge LI, Mai HU, Tongyu ZHU, Yong LIU, and Liangquan JIA
To meet the needs of monitoring CH4 gas concentration in coal mines and gas fields， a frequency modulated spectral system based on near-infrared distributed feedback lasers was built for rapid measurement of high concentration methane. The system obtained the optimal modulation frequency range and the optimal signal-toTo meet the needs of monitoring CH4 gas concentration in coal mines and gas fields， a frequency modulated spectral system based on near-infrared distributed feedback lasers was built for rapid measurement of high concentration methane. The system obtained the optimal modulation frequency range and the optimal signal-to-noise ratio of the absorption component through simulation optimization and experimentation. When the modulation frequency is 353 MHz， the power is 25.08 mW， and the effective optical path is 0.6 m， a CH4 absorption spectral line of 6 046.83 cm-1 is selected to achieve 1 Hz CH4 gas concentration detection. The amplitude of the frequency modulation absorption component signal has a linear relationship with concentration， with a linearity error of 2.88% and a linear fitting coefficient R2 of 0.997 4. For standard CH4 gas with a volume fraction of 2%， the relative error is less than 1.3‰. When the integration time is 1 s， the Allen variance is 4.032 1 ppmv. When the integration time is 347 s， the Allen variance is 0.222 8 ppmv， and the detection sensitivity is 2.28 × 10-7， which meets the measurement requirements for high-speed and highly sensitive CH4 gas. The experimental results indicate that frequency modulation spectroscopy technology has broad application prospects in low detection limit and high sensitivity gas sensing..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3537 (2024)
Sensing properties of ellipsoidal extrinsic Fabry-Perot fiber probe
Gongye LI, Shenghui SHI, Binbin LUO, Xue ZOU... and Mingfu ZHAO|Show fewer author(s)
For the detection of physical parameters in miniature complex environments， especially involving refractive index and temperature detection to meet the requirements of high sensitivity， low cost， anti-electromagnetic interference， etc. In this paper， we proposed an ellipsoidal Fabry-Perot （F-P） cavity fiber optic probeFor the detection of physical parameters in miniature complex environments， especially involving refractive index and temperature detection to meet the requirements of high sensitivity， low cost， anti-electromagnetic interference， etc. In this paper， we proposed an ellipsoidal Fabry-Perot （F-P） cavity fiber optic probe sensor with simple preparation and novel structure. Firstly， a Single Mode Fiber （SMF） with a flat cut end face was inserted into a ceramic ferrule， and UV adhesive and Graphene Oxide （GO） were sequentially fixed on the end face of the SMF to form an ellipsoidal dual F-P cavity fiber probe sensor with an end cap structure. Then the refractive index （RI） and temperature characteristics of the sensor were theoretically analyzed and experimentally investigated. The experimental results show that the interferometric spectrum of this sensor is red-shifted with the increase of RI in the range of 1.353 2 to 1.391 2， and the RI sensitivity is 119.1 nm/RIU， and the linear correlation coefficient between RI and wavelength shift is 0.99； in the range of 20~60 ℃， the temperature sensitivity of the sensor is 116 pm/℃. The probe sensor designed in this paper has the advantages of simple structure， strong anti-interference ability， low production cost， stable output spectrum and small detection area. Meanwhile， the unique physicochemical properties of GO make it have good application prospects in the field of biological detection of various antigen molecules， bacteria and viruses..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3545 (2024)
Image gray scale information adaptive error diffusion of electrowetting electronic paper
Bipeng CAI, Ting MEI, Shanling LIN, Tingyu WANG... and Zhixian LIN|Show fewer author(s)
At present， the grayscale levels of electrowetting electronic paper are not enough to meet the display needs of mainstream images. Ink reflow also affects the image quality on electrowetting electronic paper， making it difficult to distinguish details， and leading to issues such as edge blur， loss of image details， andAt present， the grayscale levels of electrowetting electronic paper are not enough to meet the display needs of mainstream images. Ink reflow also affects the image quality on electrowetting electronic paper， making it difficult to distinguish details， and leading to issues such as edge blur， loss of image details， and a strong particle sense in the displayed image. In order to solve these problems， this paper proposed an adaptive error diffusion algorithm based on characteristics of the grayscale information distribution in images. First， in order to obtain a high-fidelity grayscale reference image with 256 levels， an input color image was denoised and grayscale processed. Second， we employed sub-pixel edge detection to meticulously segment the image into edge and smooth regions. This method emphasized the edge areas while simultaneously enhancing the texture of smooth regions. This enhancement was tailored to align with the local visual characteristics of human eyes by considering the grayscale variations within the neighborhood of image. Last but not least， we adaptively adjusted the diffusion error threshold according to the grayscale information distribution characteristics of the reference image. This method reduced the error between the original reference grayscale image with 256 levels and the processed image. The experimental results showed that the images processed by this algorithm get 4.227 8 in the mean opinion score of the five-grade scoring system. The average peak signal to noise ratio reached 45.903 3 dB， marking a 16.47% to 48.89% improvement over other advanced technologies. Additionally， the average visual information fidelity reached 0.964 3， marking a 1.53% to 50.02% enhancement compared to other advanced methods. The method presented in this paper makes the edges of the images displayed on electrowetting electronic paper clearer and improves the detail and texture， thereby enhancing the readability of the electrowetting electronic paper..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3554 (2024)
Key technology research on autonomous docking control simulation system for clustered spacecraft
Chuanxiao XU, Guohua KANG, Junfeng WU, Xinyong TAO... and Yang WU|Show fewer author(s)
To address the issues of low accuracy and excessive complexity in ground-based three-degree-of-freedom air-floating platform simulations for autonomous spacecraft cluster docking experiments， we designed a continuous small-thrust， high-precision full-physical simulation system. This system was based on an embedded archTo address the issues of low accuracy and excessive complexity in ground-based three-degree-of-freedom air-floating platform simulations for autonomous spacecraft cluster docking experiments， we designed a continuous small-thrust， high-precision full-physical simulation system. This system was based on an embedded architecture and employed cascaded PID rotor propulsion control. Initially， a dynamic model for the ground-based full-physical simulation system， incorporating disturbance correction， was established. This model was developed based on the relative on-orbit dynamics of spacecraft clusters and the analysis of disturbances present in the ground simulation environment. A decoupled rotor propulsion system was then designed， which was integrated with a motion capture system， a master control host computer， and a lower control platform. The simulation spacecraft unit was developed using the μCOS operating system， and a three-degree-of-freedom motion control system was implemented based on a cascaded PID algorithm. Subsequent dual-spacecraft autonomous docking air-floating experiments were conducted to validate the system. The experimental results demonstrated that the simulation data generated by the dynamic model of the full-physical simulation system closely matched the experimental data. Specifically， the measured attitude pointing control accuracy of the ground simulation spacecraft was found to be no less than 0.1°， while the position control accuracy was no less than 1 mm. These results indicate that the developed system meets the high-precision requirements necessary for semi-physical simulation of autonomous spacecraft cluster docking. Furthermore， the successful implementation of this system provides a reference framework for the development of large-scale satellite cluster simulation systems. The approach described in this paper offers a practical solution for overcoming the challenges associated with ground-based simulations of spacecraft docking， ensuring both accuracy and system simplicity， which are crucial for advancing the study and development of spacecraft cluster operations..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3566 (2024)
High precision correction and positioning of the substrate for display ink printing
Ning AN, Hai BI, Yi LI, and Cheng XU
Aiming at the problem of large errors in the calculation of the deflection angle of the substrate and the position of the pixel pit in the display ink printing of e-paper， a high-precision substrate angle correction and location method was proposed. Firstly， a vision detection system was constructed according to the reAiming at the problem of large errors in the calculation of the deflection angle of the substrate and the position of the pixel pit in the display ink printing of e-paper， a high-precision substrate angle correction and location method was proposed. Firstly， a vision detection system was constructed according to the requirements of inkjet printing， which was used for high-resolution imaging of substrate markings. The installation angle calibration and multi-camera relative position calibration of the vision system were carried out by improving the camera's intrinsic and extrinsic parameter calibration methods and improving the microscopic imaging calibration method. Then， edge detection and contour moment feature calculation were performed on the substrate markings， the center pixel coordinates of the markings were extracted， and high-precision substrate correction was achieved using coordinate transformation and least squares line fitting. Finally， the initial pixel pit positioning was completed by indirectly using the substrate alignment marks. Experimental results show that the correction error of the substrate angle correction algorithm is within ±2″， and the pixel pit positioning error is within ±3 μm， far exceeding the accuracy requirements of ink filling. By establishing a vision detection system and designing substrate correction and positioning algorithms， high-precision and efficient printing filling of display ink were achieved， improving the yield of printing equipment， which is of great guiding significance for the localization of display ink printing and filling equipment..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3580 (2024)
Piezoelectric rotational energy harvester with indirect magnetic excitation
Shuyun WANG, Panpan LIU, Zefeng REN, Jijun ZHOU... and Zhonghua ZHANG|Show fewer author(s)
In order to meet the demand of self-power supply of remote monitoring systems under special environment， a piezoelectric rotational energy harvester with indirect magnetic excitation was proposed. The dynamic response of the system was improved by changing the height of the damping cavity， and the volume energy densityIn order to meet the demand of self-power supply of remote monitoring systems under special environment， a piezoelectric rotational energy harvester with indirect magnetic excitation was proposed. The dynamic response of the system was improved by changing the height of the damping cavity， and the volume energy density of the system was increased by coupling the piezoelectric vibrators. Through modeling and simulation analysis of harvester， the effects of exciting magnet number ratio， excited magnet distance， system equivalent stiffness and other parameters on the amplitude-magnification ratio of the system are obtained. On this basis， a prototype is designed and made， and the effects of exciting magnet number ratio， excited magnet distance， the cavity height and load resistance on the output voltage characteristics of the harvester are obtained. The results show that the number ratio of exciting magnets and the distance between exciting and excited magnets have a great influence on the maximum output voltage of the harvester， and the bandwidth of the harvester can be significantly adjusted by the cavity height， so the output voltage and effective bandwidth of harvester can be adjusted through the design of structural parameters. When other conditions are given， there is an optimal load resistance （16 kΩ） to maximize the output power （12 mW）..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3594 (2024)
Multi-frame self-supervised monocular depth estimation with multi-scale feature enhancement
Qiqi KOU, Weichen WANG, Chenggong HAN, Chen LÜ... and Yucheng JI|Show fewer author(s)
The current depth estimation networks do not sufficiently extract spatial features from images in outdoor scenes， leading to issues such as object edge distortion， blurriness， and regional pseudo-shadows in the output depth maps. To address these problems， this paper proposed a multi-frame self-supervised monocular depThe current depth estimation networks do not sufficiently extract spatial features from images in outdoor scenes， leading to issues such as object edge distortion， blurriness， and regional pseudo-shadows in the output depth maps. To address these problems， this paper proposed a multi-frame self-supervised monocular depth estimation model with multi-scale feature enhancement. Firstly， the model's encoder incorporated an activation module based on large kernel attention to enhance its ability to extract global spatial features from the input image， preserving more spatial context information. Simultaneously， a structural enhancement module was introduced that can discriminate important features across channel dimensions， enhancing the network's perception of the structural characteristics of the image. Finally， the decoder used a dynamic upsampling method instead of the traditional nearest interpolation upsampling method to restore detailed information， thereby optimizing the pseudo-shadow phenomenon in the depth map to some extent. Experimental results demonstrate that the depth estimation network proposed in this paper outperforms current mainstream algorithms in tests on the KITTI and CityScapes outdoor datasets， particularly achieving a prediction accuracy rate of 90.3% on the KITTI dataset. Visualization results also indicate that the depth maps generated by our network model have clearer and more precise edges， effectively improving the prediction accuracy of the depth estimation network..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3603 (2024)
Multi-feature cross UAV image detection algorithm under cross-layer attentional interaction
Zhihao ZHANG, Lixia DU, Yue HOU, Ziwei HAO, and Jie YIN
Aiming at the problems of complex background of aerial images， dense targets， and uneven target scale distribution in UAV traffic inspection， a multi-feature crossover under cross-layer attentional interaction （Multi-feature crossover under cross-layer attentional interaction，MCAI） UAV target detection algorithm was prAiming at the problems of complex background of aerial images， dense targets， and uneven target scale distribution in UAV traffic inspection， a multi-feature crossover under cross-layer attentional interaction （Multi-feature crossover under cross-layer attentional interaction，MCAI） UAV target detection algorithm was proposed. Firstly， an Adaptive Cross-layer Attentional Interaction （Adaptive Cross-layer Attentional Interaction，ACAI） module was designed in the backbone network part so that the model focused on the key feature regions to achieve effective screening of global key feature information， thus fading the influence of the complex background. Secondly， a deformable self-attentive encoder （Deformable Encoder， DeEncoder） was designed， which compensated for the lost target features by expanding the feature layer receptive field. Finally， in order to effectively identify tiny targets at different scales in the region， the multi-scale cross-fusion module （Multi-scale cross fusion module，MSCF） was proposed， which fused shallow spatial information and deep semantic information by combining the wavelet transform and feature representation in order to efficiently capture the fine-grained features of targets at different scales. The experimental results on the VisDrone 2019-DET， BDD-100K dataset， and LZTraffic Video dataset show that MCAI improves mAP0.5 by 3%， 2.2%， and 4.5%， respectively， compared to the RT-DETR model， which significantly improves the detection accuracy of the UAV inspection. In addition， in the cloudy and rainy scenario， the mAP0.5 of MCAI improves by 2.1% compared to the RT-DETR model， with better extreme weather robustness performance..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3616 (2024)
An optimization method for selecting common points considering position distribution and measuring precision in coordinate transformation
Zhonghe LIU, Zongchun LI, and Hua HE
The quality of coordinate transformation is influenced by geometric position distribution and measurement errors of common points. In order to improve the effectiveness of coordinate transformation， a method for selecting common points that takes into account position distribution and measurement precision is designed.The quality of coordinate transformation is influenced by geometric position distribution and measurement errors of common points. In order to improve the effectiveness of coordinate transformation， a method for selecting common points that takes into account position distribution and measurement precision is designed. Firstly， robust estimation method is adopted to process the data and eliminate gross errors. Secondly， centroid the remaining homonymous points， divide them into multiple sets of points based on different quadrants， calculate the projection length of the homonymous points on the central axis of their respective quadrants， and sort them. Proportionally select the points with larger projection lengths in the point sets as the points to be sampled. Thirdly， traverse the set of points to be sampled and extract homonymous points， calculate the conversion precision and coefficient matrix condition number of each group of points. Finally， take the combination with the highest internal compliance accuracy among the smaller number of conditions as the common points. The performance of the method is verified by simulation experiments and case analysis. The results show that compared with references ［24］ and ［28］， the proposed method has better compliance accuracy in both internal and external. The quality of coordinate transformation of this method is higher， and it has robustness..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3632 (2024)
Dermoscopic image classification based on multi-scale and three-dimensional interaction feature optimization
Di WANG, Xiaoqi LÜ, and Jing LI
Early diagnosis of skin cancer is crucial for improving patient outcomes and alleviating the burden on the healthcare system. However， the process of feature extraction in skin cancer image classification often results in information loss and challenges in simultaneously identifying independent types of features in theEarly diagnosis of skin cancer is crucial for improving patient outcomes and alleviating the burden on the healthcare system. However， the process of feature extraction in skin cancer image classification often results in information loss and challenges in simultaneously identifying independent types of features in the images. MTIFNet was proposed， which was a network that integrates three-dimensional spatial attention with information fusion. Initially， the network employed a multi-scale spatial adaptive module to extract both global and local contextual information from images during training. This module enhanced the connection between blurred pixels around lesions and the relationship in pixels at different scales. Subsequently， a three-dimensional interaction feature optimization module was introduced to facilitate connections across different dimensions， enabling the exchange and integration of information. Finally， cross-entropy loss was used to measure the difference between the predicted probability distribution and the true class distribution to optimize the accuracy of the model. The experimental results based on the ISIC 2018 and ISIC 2017 datasets indicate that the network has improved accuracy， precision， recall， and specificity to 94.32%， 91.61%， 93.00%， 98.39% and 98.57%， 98.20%， 98.47%， 99.13%， respectively. Compared to currently popular classification networks such as ResNeSt， ConvNext， and Fcanet， MTIFNet demonstrates superior capabilities in feature extraction and interaction， thereby assisting healthcare professionals in making more precise diagnostic and treatment decisions..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3644 (2024)
Airborne point cloud classification integrating edge convolution and global-local self-attention
Jingmin TU, Jin YAN, Li LI, Jian YAO... and Yanfei KANG|Show fewer author(s)
Laser point cloud classification is the basis for 3D scene understanding. In order to solve the problems of insufficient feature expression and unbalanced sample categories in the classification of large scenes of airborne point clouds， this paper proposed an airborne point cloud classification method that integrates eLaser point cloud classification is the basis for 3D scene understanding. In order to solve the problems of insufficient feature expression and unbalanced sample categories in the classification of large scenes of airborne point clouds， this paper proposed an airborne point cloud classification method that integrates edge convolution and global-local self-attention. Firstly， U-net was used as the network framework to integrate Point Transformer and edge convolution modules， so that the model could pay attention to complex ground object boundaries and texture information， and obtain local geometric features with better expressive ability. Secondly， a self-attention mechanism that integrates global context information and local structural features was innovatively proposed， and the global self-attention module tended to the information of the entire input sequence， while the local self-attention module pays more attention to the details of the local region. The combination of the two mechanisms enhanced the capture of long-distance dependence and local structure， and at the same time， the model could take into account the key features of a few categories， reduced the impact of sample category imbalance on the classification accuracy to a certain extent， and helped to improve the classification ability of the model for complex ground object relationships. Finally， the proposed method was verified on the public ISPRS-3D dataset and WHU-Urban3D dataset， and the experimental results show that the classification accuracy of the proposed method on the two datasets is 82.5% and 87.4%， respectively， which is better than that of the classical networks such as PointNet++ and Stratified Transformer and the ISPRS 3D official website competition network， which can effectively improve the classification accuracy of airborne point clouds..
Optics and Precision Engineering
- Publication Date: Dec. 25, 2024
- Vol. 32, Issue 24, 3658 (2024)
