Journals >Optics and Precision Engineering
Contents
2023
Volume: 31 Issue 20
12 Article(s)
Export citation format
Information Sciences
Indoor self-supervised monocular depth estimation based on level feature fusion
Deqiang CHENG, Huaqiang ZHANG, Qiqi KOU, Chen LÜ, and Jiansheng QIAN
Due to a high number of areas with low texture and lighting in complex indoor scenes， current self-supervised monocular depth estimation network models suffer from certain issues. These problems include imprecise depth predictions， noticeable blurriness around object edges in the predictions， and significant loss of deDue to a high number of areas with low texture and lighting in complex indoor scenes， current self-supervised monocular depth estimation network models suffer from certain issues. These problems include imprecise depth predictions， noticeable blurriness around object edges in the predictions， and significant loss of details. This paper introduces an indoor self-supervised monocular depth estimation network model based on level feature fusion. First， to enhance the visibility of poorly lit areas and address the issue of pseudo planes deteriorating the model， the Mapping-Consistent Image Enhancement module was applied to process indoor images. This module simultaneously maintained brightness consistency. Subsequently， a novel self-supervised monocular depth estimation network model that incorporates the Cross-Level Feature Adjustment module was proposed， utilizing an attention mechanism. This module effectively fused multilevel feature information from the encoder to the decoder， enhancing the network's ability to utilize feature information and reducing the semantic gap between predicted depth and true depth. Finally， the Gram Matrix Similarity Loss function was introduced based on image style features， as an additional self-supervised signal to further constrain the network model. This addition enhanced the network’s depth prediction capabilities， leading to improved accuracy. Through training and testing on NYU Depth V2 and ScanNet indoor datasets， this paper achieves a pixel accuracy rate of 81.9% and 76.0%， respectively. The experimental results also include a comparative analysis with existing main indoor self-supervised monocular depth estimation network models. The network model proposed in this paper excels in preserving object edges and details， effectively enhancing the accuracy of predicted depth..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2993 (2023)
Polarization parameter partition optimization restoration method for underwater degraded image
Ronghua LI, Changye CAI, Shenghui ZHANG, Yunhe XU, and Haotian CAO
In real water environments， common imaging problems include contrast reduction， low definition， and information attenuation. The traditional estimation method involves estimating the polarization information of the entire image. In real underwater images， the target has complex polarization characteristics， the restoraIn real water environments， common imaging problems include contrast reduction， low definition， and information attenuation. The traditional estimation method involves estimating the polarization information of the entire image. In real underwater images， the target has complex polarization characteristics， the restoration effects of some target areas are poor， and even degradation occurs. In this study， a method of polarization parameter partition optimization restoration for water-degraded images was proposed. First， the connected domain of an object with high and low polarizations was extracted after two images were processed with orthogonal polarization by block contrast enhancement and guided filtering. Based on the pixel values in the polarization image， the extraction process of high and low polarization object regions was optimized. Second， the polarization of each object was estimated， which solved the problem of incorrect estimation of complex objects in traditional global estimation methods. Finally， the image of polarization degree of backscattered light was iteratively optimized to obtain the optimal selection. Experimental results show that the subjective visual quality of the image is improved significantly. In two initial experiments， the original light intensity maps under low turbidity are compared. The measurement of enhancement by entropy （EME） value of the objective evaluation index and the contrast increases by 554% and 528% on average， respectively. In a third set of experiments， in which a comparison of the original light intensity maps in an environment of low illumination and high turbidity was conducted， the EME value and contrast are improved by 379% and 956%， respectively. Three sets of natural image quality evaluation （NIQE） indices indicate the proposed method has good performance， and a more natural image is produced. Compared with the traditional method， the proposed method can effectively restore a turbid image， increase image contrast， weaken information attenuation， and achieve a better image sharpening effect..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 3010 (2023)
Lightweight target detection network for UAV platforms
Dandan HUANG, Han GAO, Zhi LIU, Lintao YU, and Huiji WANG
A lightweight target detection network for application to unmanned aerial vehicle （UAV） platforms was proposed for solving the problems of large image-scale variation， small target size， and limited embedded computing resources on UAVs in UAV-side target detection. The network used YOLOv5 as the benchmark model. First，A lightweight target detection network for application to unmanned aerial vehicle （UAV） platforms was proposed for solving the problems of large image-scale variation， small target size， and limited embedded computing resources on UAVs in UAV-side target detection. The network used YOLOv5 as the benchmark model. First， detection branches were used to solve the problem of scale variation. Then， a small-target detection metric based on a mixture of normalized Wasserstein distance and traditional IOU was used for solving the problem of inaccurate small-target detection. In addition， a C3_FN lightweight network structure combining FasterNet and C3 was employed to reduce the computational burden of the network and make it more suitable for UAV platforms. The performance of the algorithms was tested on a simulation platform and an embedded platform using the UAV target detection dataset VisDrone. The simulation platform test results indicate that the proposed network achieves improvements of 6.6% and 4.8% in the mAP0.5 and mAP0.5-0.95 metrics， respectively， compared with a benchmark network， and the inference time is only 45.9 ms. The detection results are superior to those of mainstream UAV target detection networks. The test results for the embedded device （NVIDIA Jetson Nano） indicate that the proposed algorithm can achieve high accuracy and near real-time detection performance with limited hardware resources..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 3021 (2023)
Spatial information adaptive regulation and feature alignment for infrared methane instance segmentation
Zifen HE, Huizhu CAO, Yinhui ZHANG, and Hong ZHUANG
Conventional contact methane leak sensors suffer from a small detection range and low efficiency， but machine vision algorithms combined with non-contact infrared thermal imaging can make infrared methane instance segmentation possible at long distances and large ranges. This is a significant advantage for improving meConventional contact methane leak sensors suffer from a small detection range and low efficiency， but machine vision algorithms combined with non-contact infrared thermal imaging can make infrared methane instance segmentation possible at long distances and large ranges. This is a significant advantage for improving methane detection efficiency and ensuring personnel safety. However， the segmentation performance of infrared methane instances is limited by such problems as blurred contour and low contrast between the leaking methane gas and the background， and it can be affected by atmospheric flow factors. In response to these problems， an adaptive spatial information regulation and feature alignment network （AFNet） is proposed to segment infrared instances of methane leakage. First， to enhance the model’s feature extraction， an adaptive spatial information regulation module is proposed to endow the backbone network with adaptive weights for different scale residual blocks， which enrich the feature space extracted by the model. Second， to meet the requirements of foreground target positioning detection and contour segmentation under complex methane gas contours， a weighted bidirectional pyramid is designed to reduce the diffusion， loss of spatial location， and instance edge information in low-level features， which are caused by the top-down propagation of the feature pyramid. Finally， a prototype feature alignment module is designed to capture the semantic relationships between long-distance gas features， enriching the semantic information of the prototype and improving the quality of generated target masks to improve the methane instance segmentation accuracy. Experimental results show that the proposed AFNet model achieves AP50@95 and AP50 quantitative segmentation accuracies of 42.42% and 92.18%， which are improved by 9.79% and 6.18% compared with the original Yolact， respectively. In addition， the inference speed achieves 36.80 frames/s and meets the requirements of methane leakage segmentation. The experimental results validate the effectiveness and engineering practicality of the algorithm proposed for infrared methane leakage segmentation..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 3034 (2023)
CT and PET medical image fusion based on LL-GG-LG Net
Tao ZHOU, Xiangxiang ZHANG, Huiling LU, Qi LI, and Qianru CHENG
Multimodal medical image fusion plays a crucial role in clinical medical applications. Most of the existing methods have focused on local feature extraction， whereas global dependencies have been insufficiently explored； furthermore， interactions between global and local information have not been considered. This has lMultimodal medical image fusion plays a crucial role in clinical medical applications. Most of the existing methods have focused on local feature extraction， whereas global dependencies have been insufficiently explored； furthermore， interactions between global and local information have not been considered. This has led to difficulties in effectively addressing the complexity of patterns and the similarity between the surrounding tissue （background） and the lesion area （foreground） in terms of intensity. To address such issues， this paper proposes an LL-GG-LG Net model for PET and CT medical image fusion. Firstly， a Local-Local fusion （LL） module is proposed， which uses a two-level attention mechanism to better focus on local detailed information features. Next， a Global-Global fusion （GG） module is designed， which introduces local information into the global information by adding a residual connection mechanism to the Swin Transformer， thereby improving the Transformer's attention to local information. Subsequently， a Local-Global fusion （LG） module is proposed based on a differentiable architecture search adaptive dense fusion network， which fully captures global relationships and retains local cues， thereby effectively solving the problem of high similarity between background and focus areas. The model's effectiveness is validated using a clinical multimodal lung medical image dataset. The experimental results show that， compared to seven other methods， the proposed method objectively improves the perceptual image fusion quality evaluation indexes such as the average gradient （AG）， edge intensity （EI）， QAB/F， spatial frequency （SF）， standard deviation （SD） and information entropy （IE） edge retention by 21.5%， 11%， 4%， 13%， 9%， and 3%， respectively， on average. The model can highlight the information of the lesion areas. Moreover， the fused image structure is clear， and detailed texture information can be obtained..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 3050 (2023)
Defect detection of low-resolution ceramic substrate image based on knowledge distillation
Feng GUO, Xiaodong SUN, Qibing ZHU, Min HUANG, and Xiaoxiang XU
Ceramic substrate is a vital foundational material of electronic devices， and implementing defect detection for ceramic substrates using machine vision technology combined with deep learning strategies holds significant importance in ensuring product quality. Increasing the field of view of the imaging equipment to makCeramic substrate is a vital foundational material of electronic devices， and implementing defect detection for ceramic substrates using machine vision technology combined with deep learning strategies holds significant importance in ensuring product quality. Increasing the field of view of the imaging equipment to make simultaneous imaging of multiple ceramic substrates possible can significantly improve the detection speed of a ceramic substrate. However， it also results in decreased image resolution and subsequently reduces the accuracy of defect detection. To solve these problems， a low-resolution ceramic substrate defect automatic detection method based on knowledge distillation is proposed. The method utilizes the YOLOv5 framework to construct a teacher network and a student network. Based on the idea of knowledge distillation， high-resolution image feature information obtained by the teacher network is used to guide the training of the student network to improve the defect detection ability of the student network for low-resolution ceramic substrate images. Moreover， a feature fusion module based on the coordinate attention （CA） idea is introduced into the teacher network， enabling it to learn features that adapt to both high-resolution and low-resolution image information， thus better guiding the training of the student network. Finally， a confidence loss function based on the gradient harmonizing mechanism （GHM） is introduced to enhance the defect detection rate. Experimental results demonstrate that the proposed ceramic substrate defect detection method based on knowledge distillation achieves an average accuracy and average recall of 96.80% and 90.01%， respectively， for the detection of five types of defect-stain， foreign matter， gold edge bulge， ceramic gap， and damage-in low-resolution （224×224） input images. Compared with current mainstream object detection algorithms， the proposed algorithm achieves better detection results..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 3065 (2023)
Micro/Nano Technology and Fine Mechanics
Research of ultra-precision gear pitch artifact based on dividing mechanism with multi-tooth position
Siying LING, Yuanhang LIU, Haifeng GONG, Fengtao WANG, and Ming LING
Gear pitch artifacts are measurement standards used to calibrate the indication error of total cumulative deviation， the measurement repeatability of individual pitch deviation， and the angular positioning error of the rotary table of gear measuring instruments. A multitooth positioning claw was constructed based on thGear pitch artifacts are measurement standards used to calibrate the indication error of total cumulative deviation， the measurement repeatability of individual pitch deviation， and the angular positioning error of the rotary table of gear measuring instruments. A multitooth positioning claw was constructed based on the error-homogenization principle for the indexing mechanism of a gear grinding machine to develop high-precision gear pitch artifacts. Next， simulations for the error homogenization and indexing accuracy retention of the multitooth positioning indexing mechanism were conducted. Finally， a comparative machining experiment of the gear pitch artifacts was performed based on single-tooth and multitooth positioning indexing mechanisms. Compared to the single-tooth positioning indexing mechanism， the multitooth positioning indexing mechanism decreased the single-pitch deviation of the ground gears by 11.1%-36.4% and the total cumulative pitch deviation by 30.4%-48%. This resulted in ultraprecision grinding of a gear pitch artifact with a normal modulus mn=4 mm， a number of teeth z=30， an individual pitch deviation fp=0.6 μm， and a total accumulated tooth pitch deviation Fp=1.4 μm. The pitch deviations of the proposed gear pitch artifacts satisfy the specifications of the Chinese national standard for cylindrical gears （GB/T10095.1-2022） for Class-1 pitch tolerances （fpT=1.7 μm； FpT=5.0 μm）. This study provides a valuable reference for the ultraprecision machining of gear pitch artifacts..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2951 (2023)
Coupled hysteresis model and its inverse compensation for three-degree-of-freedom tip-tilt-piston piezoelectric stage
Jinfu SIMA, Leijie LAI, Pengzhi LI, Yu FANG, and Limin ZHU
To solve the problems of multi-axis coupling and hysteresis in a three-degree-of-freedom tip–tilt–piston piezoelectric stage， a coupled hysteresis model was designed to simultaneously characterize the coupling effect between multiple piezoelectric actuators and their own hysteresis effect. Its inverse model was used foTo solve the problems of multi-axis coupling and hysteresis in a three-degree-of-freedom tip–tilt–piston piezoelectric stage， a coupled hysteresis model was designed to simultaneously characterize the coupling effect between multiple piezoelectric actuators and their own hysteresis effect. Its inverse model was used for feedforward compensation to increase the positioning and trajectory tracking accuracies of the stage. First， the control system and kinematics model of the three-degree-of-freedom piezoelectric stage were developed， and the three-degree-of-freedom motion of the end-effector was transformed into the outputs of three piezoelectric actuators. Then， a coupled hysteresis model based on the Prandtl–Ishlinskii model was established， and the parameters of the model and its inverse model were identified. Finally， the effectiveness of the coupled hysteresis model was verified through open-loop inverse model feedforward compensation， and a compound control method combining inverse model feedforward and feedback was used for trajectory tracking control. The experimental results indicate that the inverse open-loop compensation reduced the maximum coupling displacements between the three piezoelectric actuators by >70%， confirming the effectiveness of the developed coupling hysteresis model. The maximum root mean square errors of the compound control method combined with closed-loop feedback for tracking the spatial trajectory are only 0.06 mrad and 0.42 μm， which are reduced by 72% and 87.5%， respectively， compared with those in the case where only closed-loop feedback was used， and the maximum error is reduced by at least 76%. The proposed coupled hysteresis model and its inverse compensation can eliminate the influence of coupling hysteresis in the stage， and significantly increase the positioning accuracy of the stage..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2964 (2023)
Study on the film-forming process of high-viscosity adhesive droplets by extrusion
Zheng XU, Wenxin HE, Leijie CUI, Xiaodong WANG, and Shiqin LU
During adhesive microbonding， both the adhesive thickness and the filling status can significantly affect the bond strength and creeping. In this study， the film-forming process by extrusion using a high-viscosity adhesive was examined. A typical group of tiny round parts is selected for adhesive microbonding. The simuDuring adhesive microbonding， both the adhesive thickness and the filling status can significantly affect the bond strength and creeping. In this study， the film-forming process by extrusion using a high-viscosity adhesive was examined. A typical group of tiny round parts is selected for adhesive microbonding. The simulation model describing the adhesive microbonding and the ratio of space occupancy by the adhesive， which represents the filling status， was established. The process parameters were initially determined based on the evaluated results， such as the amount of adhesive droplets. A precision adhesive microbonding assembly device was developed， and an epoxy resin adhesive was selected as the sample for adhesive microbonding experiments and analysis. Under suitable experimental conditions， the ratio of space occupancy by the adhesive increased with increasing amount of adhesive droplets， but tended to stabilize at 91.8%. In addition， adhesive thickness had an approximately linear relationship with extrusion force， and this relationship was less affected by the amount of adhesive droplets. An adhesive thickness of 25 μm was achieved as the extrusion force reached 7.92 N. The results of this study will help improve the existing adhesive microbonding process..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2986 (2023)
Modern Applied Optics
Cavity ring-down spectrometer of CO2 with ppb detection sensitivity
Ying LIU, Mai HU, Xingping WANG, Zhenyu XU... and Ruifeng KAN|Show fewer author(s)
Carbon dioxide （CO2） is the most important greenhouse gas in the atmosphere， characterized by high concentrations with minimal annual fluctuations. Therefore， high-precision monitoring of its concentration is an important link to realize the objective of the "double carbon." In this study， a CO2 gas sensing dCarbon dioxide （CO2） is the most important greenhouse gas in the atmosphere， characterized by high concentrations with minimal annual fluctuations. Therefore， high-precision monitoring of its concentration is an important link to realize the objective of the "double carbon." In this study， a CO2 gas sensing device， with detection sensitivity in the ppb level， was constructed based on continuous-wave cavity ring-down spectroscopy. The CO2 absorption line with a central wavelength of 6 251.760 cm-1 was selected. Moreover， a quartz glass Fabry-Perot resonant cavity with ultra-high fineness （>300 000） and temperature and pressure control modules with good performance were designed in the system. The changes in gas temperature and pressure in the cavity during 24 h are less than 0.07 ℃ and 15 Pa， respectively. The Allan variance result shows that the system can obtain a detection limit of 0.7×10-12 cm-1 at an optimal integration time of 303 s. For carbon dioxide， the detection limit corresponds to a minimum detectable concentration of 1.6×10-9. The linear correlation coefficient of the system's response is 0.999 94 over a wide range of CO2 concentrations. Finally， an observation of the atmospheric CO2 was conducted for 2 days with a system response time of 10 s. The results are in good agreement with the monitoring data from the commercial instrument （Picarro， G2401）， and the deviation between the two devices is less than 6‰ after excluding the interference from human exhalation. With its simple structure， low cost， and extremely high sensitivity， the system exhibits a broad application in the field of trace gas monitoring..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2921 (2023)
Comparison of deviation in aircraft casing deformation measurement based on data mapping optimization
Hui LIU, Jin LIANG, Meitu YE, Jianying GUO, and Leigang LI
A method to systematically map DIC measurement data to finite element simulation data is proposed. This addresses challenges in quantifying binocular DIC deformation measurement discrepancies in full-field during the performance testing of aircraft engine casing. Initially， two sets of point cloud data were accurately A method to systematically map DIC measurement data to finite element simulation data is proposed. This addresses challenges in quantifying binocular DIC deformation measurement discrepancies in full-field during the performance testing of aircraft engine casing. Initially， two sets of point cloud data were accurately registered using the FPFH feature and ICP algorithm， achieving precise alignment of their coordinate systems. Subsequently， a fitting neural network optimized by a genetic algorithm was employed to adjust the positions of finite element nodes. This ensured consistency in node positions between both data types， facilitating high-precision mapping from the simulation grid to the DIC grid. By implementing a point-by-point least squares strain estimation algorithm， the strain calculation techniques of both the finite element and DIC methods were aligned. Hence， finite element data that matches DIC attributes was produced， enabling estimation of full-field deformation deviations on the measured surface. The deformation comparison， particularly on the rib plate during the casing stiffness experiment， revealed a mapping accuracy of the mesh nodes better than 1×10-6 mm. Deviation images comparing simulated and DIC deformations aligned well with the deviation curve， clearly indicating the locations of DIC measurement discrepancies. This method holds significant promise for applications in the development and testing of aircraft engine casings and box-like structures..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2930 (2023)
Photoelectric inequivalence characteristics of an electric substitution radiative heat flux meter
Xin YE, Xiangyuan ZHENG, and Zhitao LUO
In non-vacuum environments， radiation heat flux meters based on the electric substitution measurement principle face challenges such as intricate photoelectric inequality and hurdles in experimental testing and correction. To enhance the meter's accuracy， the photoelectric inequivalence source of the radiant heat fIn non-vacuum environments， radiation heat flux meters based on the electric substitution measurement principle face challenges such as intricate photoelectric inequality and hurdles in experimental testing and correction. To enhance the meter's accuracy， the photoelectric inequivalence source of the radiant heat flow meter was first analyzed. Subsequently， a thermal structure model for the radiant heat flow meter was developed by combining heat transfer theory with finite element analysis. The model's validity was then ascertained via a vacuum-to-air ratio experiment. Using this finite element thermal structure model， adjustments were made to address the inequivalence in the heat transfer process. The difference between the test results of vacuum-air responsiveness of the finite element model and experimental results is 1.7%， and the inequivalence of heat transfer is 0.28%. The photoelectric inequivalent correction coefficient is 1.002 35， and the relative uncertainty is 0.29%. Hence， this approach refines the radiant heat flux meter's correction system， improves its measurement accuracy， and furnishes valuable recommendations for further optimization and enhancement..
Optics and Precision Engineering
- Publication Date: Oct. 25, 2023
- Vol. 31, Issue 20, 2943 (2023)
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20