Journals >Optics and Precision Engineering
Contents
2023
Volume: 31 Issue 17
13 Article(s)
Export citation format
Information Sciences
Partial optimal transport-based domain adaptation for hyperspectral image classification
Bilin WANG, Shengsheng WANG, and Zhe ZHANG
Hyperspectral image classification is a major task in remote sensing data processing. To solve the problem of inconsistent distribution of labeled source and unlabeled target domains， an unsupervised domain adaptive method based on partial optimal transport is proposed to achieve pixel-level classification of hyperspecHyperspectral image classification is a major task in remote sensing data processing. To solve the problem of inconsistent distribution of labeled source and unlabeled target domains， an unsupervised domain adaptive method based on partial optimal transport is proposed to achieve pixel-level classification of hyperspectral ground objects under different data distributions. Specifically， a deep convolution neural network is used to map the sample to the potential high-dimensional space， and the sample transportation scheme is established based on the partial optimal transport theory to minimize the distribution discrepancy between domains. Class-aware sampling and the mass factor adaptive adjustment strategy are used to promote the class alignment between domains and establish a global optimal transport. Experiments were conducted on two open-source hyperspectral image datasets， and the classification accuracies were compared quantitatively from the three evaluation matrices of overall accuracy （OA， %）， average accuracy （AA， %）， and Kappa （×100）. Compared with the source-only method， the improved classification accuracies with the proposed method for OA and AA were 2.21% and 2.75%， respectively， and compared with the original optimal transport， the improved accuracies were 1.71% and 2.01%， respectively. These results show that the proposed model can effectively improve pixel-level classification accuracy in hyperspectral images..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2555 (2023)
Ground point cloud segmentation based on local threshold adaptive method
Peixiang ZHANG, Qi WANG, Renjing GAO, Yang XIA, and Zhenzhong WAN
The LIDAR point cloud ground segmentation algorithm in the autonomous driving sensing module has low segmentation accuracy that requires further improvement. To address this problem， a ground point cloud segmentation algorithm is proposed based on a seed point distance threshold and road fluctuation weighted amplitude The LIDAR point cloud ground segmentation algorithm in the autonomous driving sensing module has low segmentation accuracy that requires further improvement. To address this problem， a ground point cloud segmentation algorithm is proposed based on a seed point distance threshold and road fluctuation weighted amplitude adaptive approach. Firstly， the method establishes a correlation between the selection threshold of seed points and the horizontal distance feature of the two-dimensional plane based on polar coordinate raster map division and controls the update of the seed point set through the change in horizontal distance between point clouds. Subsequently， in the process of road model fitting， the slope continuity judgment criterion is introduced to solve the stagnation problem of the slope pavement model update. Finally， the segmentation threshold equation of point clouds is established according to the change in the weighted amplitude of road surface fluctuation. This enables the achievement of adaptive threshold segmentation with respect to the distance feature of point clouds. In this paper， point cloud binary classification data processing on the open-source dataset Semantic KITTI is performed， and the performance of the algorithm is tested. The experimental results demonstrate that the ground segmentation algorithm described in this paper exhibits an improvement of 2%-4% in precision and recall when compared to existing algorithms. This substantiates the high accuracy of the algorithm proposed in this study..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2564 (2023)
Improved Retinex low-light image enhancement for stage scenes
Yuan JI, Xingyi LI, Xinde MA, and Liang LIAO
To address the problem of limited luminance dynamic range in stage scenes， this study proposes a method for low-light enhancement of the stage based on an improved Retinex algorithm. First， the low-light image of the stage scene is enhanced using the improved Retinex algorithm to obtain an overall enhanced image. Then，To address the problem of limited luminance dynamic range in stage scenes， this study proposes a method for low-light enhancement of the stage based on an improved Retinex algorithm. First， the low-light image of the stage scene is enhanced using the improved Retinex algorithm to obtain an overall enhanced image. Then， the original image is fused with the enhanced image， and background areas that are over-enhanced and do not need to be enhanced are processed to obtain the final image. The improved Retinex algorithm uses a Gauss-Laplace high-pass filter to find the reflected and illumination components， thus addressing the problem of detail loss in the reflection component. It then performs contrast and detail enhancement on the reflected component and multiplies it with the light component to produce the enhanced image. This method performs field-programable gate array （FPGA） hardware platform verification based on software platform verification. The experimental results show that， compared with other classical methods， this method yields a noticeable visual improvement， with an average increase of 57.06% in peak signal-to-noise ratio （PSNR） and 27.34% in structural similarity （SSIM） in different stage scenes. This improvement is particularly significant in stage scenes with substantial differences in brightness between light and dark areas. The images processed by this method restore the true luminance dynamic range of the stage and exhibit good natural color saturation without distortion， ensuring better image quality..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2573 (2023)
Multidimensional attention mechanism and selective feature fusion for image super-resolution reconstruction
Jian WEN, Jianfei SHAO, Jie LIU, Jianlong SHAO... and Rong YE|Show fewer author(s)
To address the problems of poor extraction of low-resolution features and blurred edges and artifacts caused by the high loss of high-frequency information in an image super-resolution reconstruction process， this paper proposes an image super-resolution reconstruction method that combines multidimensional attention anTo address the problems of poor extraction of low-resolution features and blurred edges and artifacts caused by the high loss of high-frequency information in an image super-resolution reconstruction process， this paper proposes an image super-resolution reconstruction method that combines multidimensional attention and selective feature fusion （SKFF） as an image feature extraction module. The network comprises several basic blocks and residual operations to construct the feature extraction structure of the model， the core of which is a heterogeneous group convolution block for extracting image features. The symmetric group convolution block of this module performs convolution in a parallel manner to extract the internal information between different feature channels and performs selective feature fusion. The complementary convolution block captures the missed contextual information from the null domain， input–output dimension， and kernel dimension by full-dimensional dynamic convolution （ODconv）. The features obtained after the symmetric group convolution and complementary convolution block processes are connected via a feature-enhanced residual block to remove useless information causing interference by redundancy. The rationality of the model design is demonstrated through five ablation experiments. Peak signal-to-noise ratio （PSNR） and structural similarity （SSIM） quantitative data comparison with other mainstream super-resolution reconstruction methods on the Set5， Set14， BSDS100， and Urban100 test sets are improved， especially on the Set5 dataset with an amplification factor of 3， showing a 0.06 dB improvement over the CARN-M algorithm. The experimental results demonstrate that the proposed model has better performance indexes and visual effects..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2584 (2023)
Lightweight deep global-local knowledge distillation network for hyperspectral image scene classification
Yingxu LIU, Chunyu PU, Diankun XU, Yichuan YANG, and Hong HUANG
To address the challenges of the complex spatial layouts of target scenes and inherent spatial-spectral information redundancy of HSIs， an end-to-end lightweight deep global–local knowledge distillation （LDGLKD） method is proposed herein. To explore the global sequence properties of spatial-spectral features， the visioTo address the challenges of the complex spatial layouts of target scenes and inherent spatial-spectral information redundancy of HSIs， an end-to-end lightweight deep global–local knowledge distillation （LDGLKD） method is proposed herein. To explore the global sequence properties of spatial-spectral features， the vision transformer （ViT） is used as the teacher to guide the lightweight student model for HSI scene classification. In LDGLKD， pre-trained VGG16 is selected as the student model to extract local detail information. After collaborative training of ViT and VGG16 through knowledge distillation， the teacher model transmits the learned long-range contextual information to the small-scale student model. By combining the advantages of the two models through knowledge distillation， the optimal classification accuracy of LDGLKD on the Orbita HSI scene classification dataset （OHID-SC） and hyperspectral remote sensing dataset for scene classification （HSRS） reached 91.62% and 97.96%， respectively. The experimental results revealed that the proposed LDGLKD method presented good classification performance. In addition， the OHID-SC based on the remote sensing data obtained by the Orbita Zhuhai-1 satellite could reflect the detailed information of land cover and provide data support for HSI scene classification..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2598 (2023)
Lightweight deep learning network for accurate localization of optical image components
Xiaoming NIU, Li ZENG, Fei YANG, and Guanghui HE
Precise optical image localization is crucial for improving industrial production efficiency and quality. Traditional image processing and localization methods have low accuracy and are vulnerable to environmental factors such as lighting and noise in complex scenes. Although classical deep learning networks have been Precise optical image localization is crucial for improving industrial production efficiency and quality. Traditional image processing and localization methods have low accuracy and are vulnerable to environmental factors such as lighting and noise in complex scenes. Although classical deep learning networks have been widely applied in natural-scene object detection， industrial inspection， grasping， defect detection， and other areas， directly applying pixel-level precise localization to industrial components is still challenging owing to the requirements of massive data training， complex deep learning models， and redundant and imprecise detection boxes. To address these issues， this paper proposes a lightweight deep learning network approach for pixel-level accurate localization of component optical images. The overall design of the network adopts an Encoder–Decoder architecture. The Encoder incorporates a three-level bottleneck cascade to reduce the parameter complexity of feature extraction while enhancing the network’s nonlinearity. The Encoder and Decoder perform feature layer fusion and concatenation， enabling the Encoder to obtain more high-resolution information after upsampling convolution and to reconstruct the original image details more comprehensively. Finally， the weighted Hausdorff distance is utilized to establish the relationship between the Decoder's output layer and the localization coordinates. Experimental results demonstrate that the lightweight deep learning localization network model has a parameter size of 57.4 kB， and the recognition rate for localization accuracy less than or equal to 5 pixels is greater than or equal to 99.5%. Thus， the proposed approach satisfies the requirements of high localization accuracy， high precision， and strong anti-interference capabilities for industrial component localization..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2611 (2023)
Micro/Nano Technology and Fine Mechanics
Topology optimization of electrodes for electroosmotic flow micromixer
Jianwen SUN, Jianyu ZHANG, Bowen LI, and Yongbo DENG
Electroosmotic flow is one of the commonly used driving mechanisms for microfluidics in current chip laboratory equipment， and the electrode layout plays an important role in controlling the external electric field driven by electroosmotic flow. At present， the layout of an electroosmotic flow electrode is mostly desigElectroosmotic flow is one of the commonly used driving mechanisms for microfluidics in current chip laboratory equipment， and the electrode layout plays an important role in controlling the external electric field driven by electroosmotic flow. At present， the layout of an electroosmotic flow electrode is mostly designed based on size optimization and shape optimization， making it difficult to improve the performance of microfluidic devices. To solve this problem， a topology optimization model of an electroosmotic flow electrode is established. The filter equation and threshold projection are used to control the characteristic size of the electrode structure. The adjoint sensitivity of the model is obtained by a continuous adjoint analysis method， and then the structural design variables of electrode layout are developed. Finally， the topology of an electroosmotic flow electrode is optimized. Based on this topology optimization method， the electrode layout of an electroosmotic flow micromixer was designed， and the factors affecting its mixing effect were analyzed. The results show that the mixing evaluation index of the electroosmotic flow micromixer reaches 0.047， which can completely mix two different concentration solutions. The good mixing performance of the micromixer verifies the effectiveness of the proposed electroosmotic flow method for electrode topology optimization..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2515 (2023)
Test and evaluation of long-endurance rotary modulation inertial navigation system
Xudong YU, Ding LI, Lifu GAO, and Wen LEI
Prior to formally equipping an inertial navigation system （INS）， its navigation accuracy must be evaluated based on the national military standards. Owing to the continuous accuracy improvements of the rotary modulation INS， the duration of a single-navigation experimental cycle is increasing， thereby causing challengePrior to formally equipping an inertial navigation system （INS）， its navigation accuracy must be evaluated based on the national military standards. Owing to the continuous accuracy improvements of the rotary modulation INS， the duration of a single-navigation experimental cycle is increasing， thereby causing challenges to test and evaluate the INS system. Thus， to evaluate the navigation accuracy of a long-endurance rotary modulation INS， first， this study introduces an integrated rotary modulation scheme with self-calibration， self-alignment， and navigation functions that unify the self-calibration and self-alignment state of the INS and the rotation path in the navigation process. Subsequently， this study proposes a long-period navigation test method based on repeated samples for long-endurance rotary modulation INS to improve all required navigation testing and evaluation tasks. Finally， the study obtains and uses 30-day long-endurance navigation laboratory-test data to test and verify the maximum position error （0.71， normalized， which is equivalent to the statistical results of independent voyages）. Furthermore， the effectiveness and feasibility of the proposed method are verified using a vehicle dynamics test. The results show that the proposed method can significantly shorten the test cycle. Thus， this study provides an effective approach for developing and evaluating long-endurance rotary modulation INS..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2525 (2023)
Trajectory planning scheme for ultra-precision system
Ailin LI, and Jing LI
As equipment used for IC manufacturing must meet a wide range of complicated requirements， improving the efficiency and accuracy of multi-degree-of-freedom motion positioning systems， which are widely used in the design of various stages of IC manufacturing， has become a contentious research topic. Thus， this study proAs equipment used for IC manufacturing must meet a wide range of complicated requirements， improving the efficiency and accuracy of multi-degree-of-freedom motion positioning systems， which are widely used in the design of various stages of IC manufacturing， has become a contentious research topic. Thus， this study proposes a critical trajectory-planning scheme for application in an ultra-precise and multi-motion stage. The goal of the trajectory planning is to control nanometer-level accuracy while improving technical efficiency in fine-tuning. The optimization trajectory model is based on the practical situation and the requirements of the real processes. A differential evolutionary approach is used to modify the controller setting， and a reference trajectory is derived using an iterative Monte Carlo approach based on the actual tracking performance of the controller as the optimization target. Simulations and experiments are performed using a physical system， and the data analysis demonstrates a fast convergence of the tracking errors， with approximately 90% reduction in the level of parameter fine-tuning while maintaining the repeat positioning precision of the stage below ±5 nm/3σ. The experimental results indicate that the proposed scheme can enhance the working efficiency of a motion system towards its required accuracy while maintaining a high level of positioning precision..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2534 (2023)
Calibration of piezoelectric three-dimensional force sensor considering angle deviation
Jun ZHANG, Yuhe WANG, Zongjin REN, Wenju SUN, and Jiale CAI
As the final step in the fabrication of a force sensor， the sensor is calibrated to obtain its true performance parameters. Therefore， the accuracy of a force sensor is determined by its calibration accuracy. Generally， calibration is an ideal process， but angle deviations can reduce the calibration accuracy. To achievAs the final step in the fabrication of a force sensor， the sensor is calibrated to obtain its true performance parameters. Therefore， the accuracy of a force sensor is determined by its calibration accuracy. Generally， calibration is an ideal process， but angle deviations can reduce the calibration accuracy. To achieve high-precision calibration of piezoelectric three-dimensional force sensors， a calibration angle-deviation factor is introduced into the original calibration model. For factors such as uneven surface， sensor angle， and deflection of force source， calibration model improvement， parameter acquisition， and decoupling effect analyses are performed. First， the angle-deviation factors in sensor calibration are analyzed. A dial indicator and an inclinometer are used for measurement. Second， considering the unknown parameters， an appropriate objective function and constraint conditions are established. Based on the optimization algorithm， model parameters are solved and sensor calibration to remove calibration angle deviation is realized. Finally， two models are used to decouple the experimental data. The effectiveness of the new model is proven from the perspective of the mean and extreme values of the decoupling error. Results show that compared to the old model， the mean value of decoupling error in most directions decreases by approximately 80%， and the extreme value decreases by 20%-49%， thereby verifying the effectiveness of the proposed method. On this basis， the influence of angle deviation on sensor coupling characteristics is analyzed， and the coupling ratio introduced by each link deviation is obtained. As the calibration angle-deviation factor accounts for over 70% of partial coupling coefficients， the influence of angle deviation cannot be ignored and should be controlled..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2546 (2023)
Modern Applied Optics
Optical properties and quantum efficiency of multilayer complicated GaAs-based photocathode
Cheng FENG, Jian LIU, Yijun ZHANG, and Yunsheng QIAN
To comprehensively assess the impact of these structures on photoelectric emission performance， we design and cultivate photocathode samples featuring both configurations， subsequently comparing their photoemission performance. Optical properties and the quantum efficiency model are ascertained through the matrix methoTo comprehensively assess the impact of these structures on photoelectric emission performance， we design and cultivate photocathode samples featuring both configurations， subsequently comparing their photoemission performance. Optical properties and the quantum efficiency model are ascertained through the matrix method of thin-film optics and one-dimensional continuity equations， respectively. We examine and analyze the effects of varying the thickness of the emission layer， altering the buffer layer， and adjusting the Al component within the buffer layer on the optical properties and quantum efficiencies of these two photocathode structures via simulations. These two structures exert distinct influences on photoelectric emission performance due to their disparate mechanisms. Consequently， their effects on photoemission performance exhibit substantial differences. The photocathode with the graded bandgap structure improves the photoelectric emission performance by introducing a built-in electric field and reducing interface recombination. Conversely， the DBR structure augments the photoelectric emission by forming a Fabry Robb resonant cavity， so that incident light of a specific wavelength can be reflected back and forth in the resonant cavity and absorbed many times. The results of an activation experiment indicate that the emission efficiency of DBR structures exceeds that of graded bandgap structures. Notably， higher emission efficiency peak values are obtained at the wavelengths of 755， 808， and 880 nm， which can be improved by 37.5%， 38.9%， and 47.0%， respectively. Furthermore， the quantum efficiency curves are well fitted using the derived model， confirming the importance of optical performance parameters and the model’s validity..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2483 (2023)
Design and performance analysis of high-voltage drive system for thousand-element adaptive optics
Meng SHAO, Hongwen LI, Jianli WANG, Leqiang YANG... and Yongting DENG|Show fewer author(s)
Adaptive optics technology can correct the wavefront error caused by atmospheric turbulence in real time and is the key technology to achieve high-resolution imaging of large ground-based telescopes. As the diameter of a telescope increases， the number of correction elements in the adaptive optics system can reach the Adaptive optics technology can correct the wavefront error caused by atmospheric turbulence in real time and is the key technology to achieve high-resolution imaging of large ground-based telescopes. As the diameter of a telescope increases， the number of correction elements in the adaptive optics system can reach the order of one thousand. This paper first establishes the equivalent model of the control link of the adaptive optics system from the perspective of automatic control and analyzes the influence of the system delay on the performance of the control link. Then， it discusses the design of the high-voltage drive system in the control link of adaptive optics， analyzes the demand of the adaptive optics system for the closed-loop bandwidth of the high-voltage amplifier from the perspective of automatic control， and provides the analysis results. Finally， it describes the integration and test of the high-voltage drive system of the thousand-element adaptive optics system. The experimental results show that the designed high-voltage amplifier can achieve 120 V output， and the bandwidth of -3 dB can reach 5 000 Hz. After the integration of the designed high-voltage drive system， the turbulence screen is used to simulate the equivalent 60 Hz Greenwood frequency， and the corrected mean residual error of the wavefront is 0.16λ. The designed high-voltage drive system can achieve calibration control of the thousand-element piezoelectric deformable mirror..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2493 (2023)
Gating characteristics of GaAs photocathode image intensifier
Dong LI, Kaixiang YANG, Liang SHENG, Yang LI... and Mei ZHANG|Show fewer author(s)
Considering the spacetime characteristics of the GaAs photocathode image intensifier in ns-level gated imaging， this study undertakes a theoretical simulation and experimental validation. For theoretical simulation， the radial RLC transmission model of the photocathode is enhanced by incorporating transmission line impConsidering the spacetime characteristics of the GaAs photocathode image intensifier in ns-level gated imaging， this study undertakes a theoretical simulation and experimental validation. For theoretical simulation， the radial RLC transmission model of the photocathode is enhanced by incorporating transmission line impedance. This refinement enables a more accurate description of the optical shutter's behavior during the gating process. Experimental evidence confirms that removing the anti-ion feedback film enhances the optical shutter， aligning it closely with the electric shutter. Specifically， when the driving electric pulse width is 17.7 ns， the difference between the optical shutter width and the electric shutter width is merely 1.1 ns. For experimental validation， a spatial dispersion model of photoelectrons， driven by a segmented linear shutter pulse voltage after the first close attachment， is established using the Monte Carlo simulation method. Simulation outcomes indicate that the spatial resolution degradation of the GaAs photocathode in gating imaging is inferior to that of the S20 photocathode. At a spatial resolution of 20-line pairs per millimeter （lp/mm）， GaAs maintains 80% of its static spatial resolution， whereas the corresponding figure for the S20 photocathode is less than 70%. Notably， the theoretical simulation aligns seamlessly with the experimental results， affirming the applicability of the model for analyzing and optimizing image intensifier structural parameters. This model serves as a foundational framework for enhancing gating imaging performance..
Optics and Precision Engineering
- Publication Date: Sep. 10, 2023
- Vol. 31, Issue 17, 2505 (2023)
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20