Latest Issue
Issue 13 ， 2023
- Abstract：In order to achieve ultraviolet detection without power supply， a self-powered ultraviolet detector based on WO3 nanosheets is constructed. The uniform WO3 nanosheets have been successfully synthesized on fluorine-doped tin oxide substrate via a facile hydrothermal method. Then， the morphology， composition， and structure of the samples are systematically characterized， which indicates that WO3 nanosheets are monoclinic with an average length of 2 μm and a mean thickness of 200 nm. Furthermore， the ultraviolet detector based on WO3 nanosheets has been fabricated and measured without bias voltage for its photodetection performances， which reveals high values of photocurrent about 171 μA， rapid response with rise time around 25.7 ms and decay time around 38.7 ms， outstanding stability and self-powered property under the cycled irradiation of 365 nm ultraviolet light.Keywords：ultraviolet detector;WO3;Nanosheets;hydrothermal method
- 119
- |
- 23
- |
- 0
Published：2023-07-17 - Abstract：It is important to know the calorific value of coal in real time for adjusting the ratio of air to powder of power plant boilers and increasing the coal combustion efficiency. Currently， power production is in urgent need of a rapid， highly stable calorific-value detection method. Therefore， this paper innovatively proposes a highly stable detection method for the coal calorific value using the combination of near-infrared spectroscopy （NIRS） and X-ray fluorescence （XRF） spectroscopy， which significantly improves the measurement repeatability of the coal calorific value by combining the advantages of NIRS for highly stable detection of organic groups positively related to the calorific value in coal and XRF spectroscopy for ash-forming elements negatively related to the calorific value. The spectral preprocessing method used in this study involves fusing the two sets of spectra as input variables for partial least squares regression （PLSR） for preliminary modeling of the full spectrum， selecting the effective bands in the NIRS spectrum according to the regression coefficients， and then fusing them with the ash-forming element XRF spectra for normalization. All the data of the preprocessed fused spectra are used as input variables to model the coal calorific value using PLSR. The experimental results indicated that the linear correlation coefficient （R2） of the present NIRS-XRF coupled method for the prediction of the calorific value of the calibration set coal samples was 0.995， and the minimum root-mean-square error， average relative error， and standard deviation for the prediction of the calorific values of the validation-set coal samples were 0.24 MJ/kg， 0.61%， and 0.05 MJ/kg， respectively. The measurement repeatability fully satisfied the requirement of <0.12 MJ/kg based on the national standard. The proposed highly stable detection method combining NIRS and XRF spectroscopy for the coal calorific value is expected to be popularized and applied in high carbon industries such as thermal power generation， the coal chemical industry， metallurgy， cement， coking， etc.， to help China achieve carbon neutrality on schedule.Keywords：near infrared spectroscopy （NIRS）;X-ray fluorescence （XRF）;spectral fusion;coal calorific value;high-stability analysis
- 54
- |
- 9
- |
- 0
Published：2023-07-17 - Abstract：To measure the bump height of chip packaging， a measurement system based on white-light triangulation was established， and a method for measuring the bump height in a complex background was proposed. First， the bump-height measurement system was built according to the proposed measurement model. Next， the U-Net deep learning model was used to segment the light stripe image， and the grayscale barycenter method and interpolation method were combined for extracting the complete light stripe center to overcome the interference of the complex background. Subsequently， the measurement system was calibrated to determine the corresponding relationship between the pixel value and the actual height value. Finally， the height of the bump was determined via function fitting. The experimental results indicate that the average repeatability of bump-height measurement is 0.21 μm， with a standard deviation of 0.095 μm； the average measurement error is ⁃0.04 μm， with a standard deviation is 0.408 μm. In this study， bump-height measurement with high accuracy and robustness in a complex background was realized， which can satisfy the requirements of online detection of wafer bump coplanarity in chip packaging.Keywords：chip detection;deep learning;light stripe center;bump measurement
- 100
- |
- 20
- |
- 0
Published：2023-07-17 - Abstract：To solve the problem of the reduction in the position accuracy of the laser spot centroid caused by laser random speckle in the high-precision visual positioning measurement system， the speckle mechanism and suppression methods were studied. First， the causes of laser random speckle in the laser spot imaging process were analyzed， and the expression for the total wave disturbance at any point in space was derived. Then， a model for the relationships among the light absorption， roughness of the imaging medium， surface outgoing light signal-to-noise ratio， and random speckle intensity was established. It is concluded that increasing the roughness and light absorption of the imaging medium can increase the signal-to-noise ratio of the outgoing light and thus reduce the random speckle intensity. Finally， comparison experiments were performed to obtain the trend chart of the stability of the spot centroid position with changes in the physical characteristics of the imaging medium， and the results validated the relationship model. The experimental results indicate that when the laser power is 50 mW， the imaging medium had strong light absorption， and the imaging target surface had moderate roughness and thickness， the stability of laser spot centroid extraction is within 0.04 pixels. Laser speckle is suppressed， the stability of imaging spot coordinate extraction is improved， and the accuracy of spot centroid positioning is increased.Keywords：Laser spot;dynamic speckle;scattering of light;molecular motion theory;microfacet model
- 78
- |
- 16
- |
- 0
Published：2023-07-17
Modern Applied Optics
- Abstract：The cutting characteristics of the polymer determine the processing quality of the microstructure. Considering photoresist SU8 as the representative polymer in this research， the cutting characteristics of the photoresist mask were studied through experimental analysis and simulations. The stress-strain relationship of SU8 was analyzed via the nanoindentation method， and a cutting simulation model of SU8 based on the energy method was established. Next， the cutting characteristics of SU8 under different cutting parameters were simulated using AdvantEdge FEM. Finally， an experiment involving ultra-precision machining of SU8 was performed. According to the simulation and test results， the effects of the cutting parameters and tool rake angle on the surface quality were analyzed， and the cutting parameters of SU8 were optimized. The results demonstrat that the surface roughness decreases with an increase in the cutting speed and increases with the feed rate and depth of cut. Within the scope of the experimental conditions， when the cutting speed is 2.09 m/s， the feed speed is 1 mm/min， the depth of cut is 2 μm， and the tool rake angle is 0°； the surface roughness Ra of the photoresist reaches the optimum value of 7.4 nm， and there are no microfractures. Finally， according to the test and simulation results， the processing parameters were optimized， and a microlens array mask structure was fabricated on SU8 with high precision.Keywords：ultra-precision cutting;mask processing;cutting simulation;photoresist SU8;microlens array
- 82
- |
- 6
- |
- 0
Published：2023-07-17 - Abstract：For a certain type of airborne optical electronic load with frame-division step imaging， the system undergoes a sudden change in the given position value and slight shaking of the camera body when the motor starts and stops. The system takes so long that the steady-state setting time exceeds the time-sequence control requirements， which affects the image quality. This study proposes a control method called cascade dynamic hybrid amplitude limitation. On the basis of cascade control with primary and subsidiary loops， the system realizes the adaptation of the sudden jump in the given step angle value and reduces the influence of slight body shaking via dynamic programming of voltage limits and speed limits according to the photoperiodic starting signal and position-speed control signal. Finally， the system satisfies the time-sequence control requirements of an airborne optical electronic load. Experiments and dynamic imaging tests indicate that when cascade dynamic hybrid amplitude limitation control is adopted， the system realizes fast and steady-state performance of step position control during the imaging and back periods. The steady-state setting time of position control during the imaging and back periods is reduced by 26.56% and 18.47%， respectively. This method satisfies the time-sequence control requirements of an airborne optical electronic load and improved the rolling direction image motion compensation effect.Keywords：airborne optical electronic load;aerial camera;amplitude limitation;anti-windup;cascade control
- 40
- |
- 8
- |
- 0
Published：2023-07-17 - Abstract：The micro-driving stage is the key component of high-precision positioning. The traditional work stage has few degrees of freedom （DOFs） and a low resolution； it struggles to adapt to the development needs of precision measurement and other fields. Thus， this paper presents a 6-DOF high-precision micro-driving stage and its experimental performance. The overall structure is hollow and is based on the series-parallel hybrid drive method. The movement of the six DOFs is distributed reasonably in the translational， rotating， and supporting plates. We also designed and developed a motion-control system for the micro-driving stage and motion-control software that can be used with it. Experimental results indicated that the linear displacement ranges of the X-， Y-， and Z-axis are better than 20， 20， and 37 μm， respectively， and the angular range is better than 39"， 33"， 27" in the pitch， roll， and yaw directions， respectively. The linear displacement resolution is better than 0.7 nm， and the angular displacement resolution is better than 0.1". Compared with the traditional work stage， the proposed 6-DOF micro-driving stage has the advantages of more DOFs and a higher resolution； thus， it is expected to be widely used in ultra-precision machining and lithography equipment.Keywords：Micro-driving stage;resolution with sub-nano;6-DOF;motion control system
- 82
- |
- 15
- |
- 0
Published：2023-07-17 - Abstract：Mechanical ruling is the main processing method for gratings with a large blaze angle and high diffraction order， such as echelle. The large blaze angle of echelle makes its ruling process difficult. When the angle between the ruling tool edge direction and the ruling direction exceeds the error or when the ruling tool is designed without force balance， the tool undergoes a large horizontal torque during the ruling process， which can easily lead to tool chatter， affecting the quality of the gratings. To suppress the tool chatter and improve the stability of the grating ruling process， the chatter suppression performance of the grating ruling tool elastic support mechanism is evaluated in this study， and an innovative double-layer parallel hinge tool carrier is proposed. The structural stiffness and modal characteristics of the new and old tool carriers are compared and analyzed through finite-element simulation. The chatter state of the two tool holders is measured using an acceleration sensor， and the chatter suppression effects of the two tool holders are compared. The results indicate that when the double-layer parallel spring tool carrier is subjected to the ruling torque， the deformation at the tool clamping position and the flexure hinge is reduced by 36% and 24%， respectively， compared with the previous structure， and the first-order mode is increased by a factor of 2.8. The amplitude of the flutter signal on both sides of the baseline is reduced by 13.6% and 22.5%， respectively， in the time domain. The test results confirm that the proposed mechanism is advantageous for echelle ruling. The proposed tool carrier optimization design method and flutter suppression technology provide new theoretical guidance for grating ruling technology.Keywords：mechanical ruling method;ruling tool carrier;chatter signal;echelle grating
- 69
- |
- 13
- |
- 0
Published：2023-07-17
Micro/Nano Technology and Fine Mechanics
- Abstract：For solving the problem of spectral shift between the source domain and target domain in cross-scene hyperspectral remote sensing image classification， this study proposes a cross-scene hyperspectral image classification model combining spatial-spectral domain adaptation and Xtreme Gradient Boosting （XGBoost）. First， the Depth Over Parametric Convolution Model （DOCM） and Large Kernel Attention （LKA） was combined to form a spatial-spectral attention model and extract the spatial-spectral features of the source domain. Next， the same spatialspectral attention model was used to extract features from the target domain， and the discriminator was used to adapt to the confrontation domain to reduce the spectral shift between the source and target domains. Second， the feature extractor of the target domain was adapted to the supervised domain through a small amount of labeled data in the target domain such that the feature extractor of the target domain can learn the true distribution of the target domain and map the features of the source and target domains to form a similar spatial distribution and complete the clustering domain adaptation. Finally， the ensemble classifier XGBoost was used to classify hyperspectral images to further improve the training speed and confidence of the model. Experimental results for the Pavia and Indiana hyperspectral datasets indicate that the overall classification accuracy of this algorithm reaches 91.62% and 65.98%， respectively. Compared with other cross-scene hyperspectral image classification models， the proposed model has a higher classification accuracy.Keywords：hyperspectral images;domain adaptation;large kernel attention;XGBoost
- 75
- |
- 13
- |
- 0
Published：2023-07-17 - Abstract：Considering that the existing remote-sensing ship datasets consist entirely of cropped images， the detection effect of the detection algorithm trained on the datasets is poor when it is directly applied to satellite images of the original scale. In this study， a multispectral satellite ship dataset MMShip with four bands of visible and near-infrared （NIR） light was established. The dataset includes both the original-scale data of satellite images and cut small-scale ship data. Owing to the introduction of multi-band information， this dataset compensates for the shortcoming that most of the existing datasets contain visible images， which are easily affected by illumination conditions. Sentinel-2 satellite images with cloud cover of <3 in the oceans worldwide were downloaded. After atmospheric correction， only four bands—red， green， blue， and NIR—with a 10-m resolution were selected， and the images containing ships were screened by scene. Next， the screened images were divided into a size of 512 × 512 such that the divided images do not overlap， and the images that did not contain the ship target were eliminated. The LabelImage software was used to label the small-scale data with a horizontal frame， and then the labeled data were converted to the original scale to obtain the labeling information under the original scale. Finally， several typical detection algorithms were used to perform visible-light， near-infrared， and multispectral comparison experiments on the altered MMShip small-scale dataset. In this study， a multispectral satellite ship target dataset covering different scenes was constructed， which included 497 original scale-labeled data and 5 016 groups of cropped ship target images. The contrast experiment confirmed that the addition of near-infrared band information can increase the accuracy of the ship target detection algorithm. The developed multispectral ship dataset MMShip can be applied to research on algorithms for multispectral ship target detection at the satellite-image and ordinary-image scales.Keywords：multispectral remote sensing;ship dataset;ship detection;sentinel-2
- 48
- |
- 11
- |
- 0
Published：2023-07-17 - Abstract：Multiple entropy thresholding （MET） increases exponentially with an increase in the number of thresholds K. Related optimization strategies exhibit low accuracy and stability with the segmented aggregate images lacking considerable feature information such as surface roughness and edges. To overcome these problems， an automatic image segmentation model based on a chaotic sparrow search algorithm （SSA） was developed to optimize MET. SSA is a newer intelligent optimization algorithm. To enhance the global optimization capability and robustness of SSA， a logistic map is added to the uniform sparrow distribution at the time of population position initialization， an expansion parameter is applied to expand the global search， and temporal local stagnation is avoided by range-control elite mutation jumps. This algorithm is called logistic SSA （LSSA） and can improve the solution quality without reducing convergence speed. LSSA is used for the automatic selection of MET parameters， with the Renyi entropy， symmetric-cross entropy， and Kapur entropy as objective functions to quickly determine the correct thresholds. In this study， image segmentation and algorithm comparison experiments are conducted on aggregate images with different characteristics. The effectiveness of LSSA-MET was demonstrated by comparing six types of combined algorithms with the fuzzy C-means （FCM） algorithm. The proposed algorithm maintains a relatively high speed with an increase in K， taking 1.532 s to split an image on average even when K=6. Among the variousm entropies， LSSA-Renyi entropy performed the best， achieving 29.92%， 10.67%， and 5.16% accuracy improvements in peak signal-to-noise ratio （PSNR）， structural similarity （SSIM）， and feature similarity （FSIM）， respectively， thereby effectively retaining the aggregate surface texture and edge characteristics while achieving the optimum balance between precision and speed.Keywords：image segmentation;aggregate;multiple entropy thresholding;optimization;sparrow search algorithm;logistic map
- 39
- |
- 8
- |
- 0
Published：2023-07-17 - Abstract：The construction and operation of subways causes different degrees of impact on the ground， particularly the construction of subways on islands， which have complex topographic and geomorphological environments， unstable soil layers， and ground deformation that is difficult to analyze. The objective of this study was to solve these problems. In this study， using the 38-view image data of Sentinel-1 and tunnel point-cloud data collected independently， the permanent scatterer synthetic aperture radar measurement technology， differential interferometry short baseline set time-series analysis technology， and LiDAR technology were employed to study the area along Metro Line 2 within Xiamen Island. The accuracy of the results of these two techniques based on synthetic aperture radar were compared with level data to verify their accuracy. Finally， the results of deformation were compared with the results of the LiDAR technique to obtain point-cloud data by scanning the tunnel inside the subway. The experimental results show that the deformation rate based on PS-InSAR technology ranged from -24.21 to 24.19 mm/y， whereas the deformation rate based on SBAS-InSAR technology ranged from -22.86 to 33.79 mm/y； the two methods were essentially identical in the process of deformation monitoring. The comparison with the level measurement results indicate that there was an error between the two. The error was concentrated within ±7 mm and was approximately 13 mm for some of the distances from the level. Meanwhile， the settlement range of the underground tunnel was -43.4 to 104.1 mm， with a medium error of 28.44 mm， and the result differed significantly from those for the ground deformation. After several verifications， it was confirmed that the underground rail transit tunnel is less affected by the ground deformation.Keywords：lidar;synthetic aperture radar;surface deformation;tunnel deformation
- 45
- |
- 12
- |
- 0
Published：2023-07-17 - Abstract：Detecting indoor instance objects is useful for various applications. Traditional deep-learning methods require a large number of labeled samples for network training， making them time-consuming and labor-intensive. To address this problem， SVD-RCNN—a semi-supervised instance object detection network based on singular value decomposition （SVD） and co-training—is proposed. First， key samples are selected for manual labeling to pre-train SVD-RCNN， to ensure that it acquires more prior knowledge. Second， a convergence， decomposition， and finetuning strategy based on SVD is used to obtain two detectors with strong independence in SVD-RCNN to satisfy the requirements of co-training. Finally， an adaptive self-labeling strategy is used to obtain high-quality self-labeling and detection results. The method was tested on multiple indoor instance datasets. On the GMU dataset， it achieved a mean average precision of 79.3% with 199 manually labeled samples. This was only 2% lower than that （81.3%） of Faster RCNN with fully supervised learning， which required labeling 3 851 samples. Ablation studies and a series of experiments confirmed the effectiveness and universality of the method. The results indicated that the method only needs to manually label 5% of the training data to achieve instance-level detection accuracy comparable to that of fully supervised learning； thus， it is suitable for applications in which intelligent robots must efficiently identify different instance objects.Keywords：computer vision;object detection;semi-supervised learning;self-labeling
- 56
- |
- 13
- |
- 0
Published：2023-07-17
Information Sciences
0