Journals >Optics and Precision Engineering
Contents
2023
Volume: 31 Issue 14
13 Article(s)
Export citation format
Information Sciences
Generate adversarial network for super-resolution reconstruction of remote sensing images by fusing edge enhancement and non-local modules
Jie LIU, Ruo QI, and Ke HAN
To address the serious noise pollution in the process of image remote sensing and the existence of object edge blur and artifacts in the super-resolution reconstructed image， this study proposes a remote sensing image super-resolution algorithm called edge-enhanced and non-local modules generative adversarial network （To address the serious noise pollution in the process of image remote sensing and the existence of object edge blur and artifacts in the super-resolution reconstructed image， this study proposes a remote sensing image super-resolution algorithm called edge-enhanced and non-local modules generative adversarial network （ENGAN）. To make the image edge details clearer， the proposed algorithm integrated an image edge enhancement module. To further expand the receptive field of the model and enhance the edge noise removal， the Mask branch in the edge enhancement module was simultaneously improved. The use of the intrinsic feature correlation of images further improved the reconstruction performance of the network. In this study， comparison experiments of multiple algorithms were performed on two remote sensing image datasets， UCAS-AOD and NWPU VHR-10. The proposed method showed improvement in multiple evaluation indicators. Taking degradation type IV as an example， the 4x super-resolution SSIM was increased by 0.068， PSNR increased by 1.400 dB， and RMSE reduced by 12.5% compared with the deep-blind super-resolution degradation model. Moreover， the reconstructed remote sensing image can obtain better ground target detection results than the original image..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2080 (2023)
REC-ResNet： Feature enhancement model for COVID-19 aided diagnosis
Tao ZHOU, Yuncan LIU, Senbao HOU, Xinyu YE, and Huiling LU
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2093 (2023)
Low-light image enhancement algorithm based on multi-channel fusion attention network
Qingjiang CHEN, and Yuan GU
Low-light images have low brightness， low contrast， and color distortion， and most existing enhancement algorithms do not deal with different channels differently， which is not conducive to the extraction of multi-level features. Therefore， this study proposes a low-light image enhancement algorithm based on a multi-chLow-light images have low brightness， low contrast， and color distortion， and most existing enhancement algorithms do not deal with different channels differently， which is not conducive to the extraction of multi-level features. Therefore， this study proposes a low-light image enhancement algorithm based on a multi-channel fusion attention network. Firstly， we introduced octave convolution （OctConv） into the residual structure after channel splitting and propose a multi-level feature extraction module. Secondly， we proposed a cross-scale feature attention module using an attention mechanism and cross-residual structure. Thirdly， we obtained multi-level information by stacking modules with different sizes and channels. Finally， we performed feature fusion in the channel dimension and obtained the final output through the reconstruction module. The experimental results showed that compared with the RISSNet algorithm， the peak signal-to-noise ratio and structural similarity of real images were improved from 27.001 6 dB and 0.889 2 to 27.978 1 dB and 0.925 5， respectively. The proposed algorithm achieved the best results in four objective evaluation indicators： peak signal-to-noise ratio， structural similarity， mean squared error， and visual information fidelity. The algorithm can effectively improve the brightness and contrast of low-light images with well-maintained image textures and colors..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2111 (2023)
Weather recognition combining improved ConvNeXt models with knowledge distillation
Libo LIU, Siyu XI, and Zhen DENG
A weather recognition method combining an improved ConvNeXt network and knowledge distillation is proposed to improve the accuracy of weather recognition in complex traffic scenes while achieving network lightweighting. Firstly， the ConvNeXt~~F network was constructed， and the SimAm attention mechanism was added after A weather recognition method combining an improved ConvNeXt network and knowledge distillation is proposed to improve the accuracy of weather recognition in complex traffic scenes while achieving network lightweighting. Firstly， the ConvNeXt_F network was constructed， and the SimAm attention mechanism was added after each set of Block feature extraction of the ConvNeXt network to correct the weights of the extracted deep features and strengthen the ability to capture discriminative weather features. Secondly， during the network training， equalized focal loss （EFL） and mutual-channel loss （MCL） were aggregated as the total loss function by using the average occupancy ratio， eliminating the effect caused by data imbalance using EFL and reducing the difference of local detail features under similar weather using MCL. Finally， the knowledge distillation technique was used to migrate the weather classification knowledge from the ConvNeXt_F network to the lightweight MobileNetV3 network， which has a marginal loss of accuracy but significant reduction in the number of network parameters. The experimental results showed that compared with other algorithms， the proposed method achieved 96.22% and 84.8% accuracy on the weather-traffic dataset of Ningxia expressway and publicly-available natural weather dataset RSCM2017， respectively； the FPSs were 157.6 Hz and 137.6 Hz and FLOPs and Params were 0.06 G and 2.54 M. Compared with the original network， the recognition accuracy， speed， and lightness of the network were improved， making it better applicable to practical scenarios with limited storage and computational power..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2123 (2023)
Image reconstruction based on deep compressive sensing combined with global and local features
Yuanhong ZHONG, Qianfeng XU, Yujie ZHOU, and Shanshan WANG
Effectively restoring the original signal with high probability and high quality from a very small number of measured values is the core issue of compressive sensing for image reconstruction. Researchers have successively proposed traditional and deep learning-based compressive sensing image reconstruction algorithms. Effectively restoring the original signal with high probability and high quality from a very small number of measured values is the core issue of compressive sensing for image reconstruction. Researchers have successively proposed traditional and deep learning-based compressive sensing image reconstruction algorithms. The traditional algorithms are based on mathematical derivation. Although they are comprehensible， their reconstruction quality is relatively poor. On the contrary， deep learning-based algorithms have relatively high reconstruction quality， but they cannot guarantee intelligibility. Inspired by filter flow， this study proposes a global-to-local compressive sensing image reconstruction model called G2LNet， which performs compressed sampling and initial reconstruction processes with convolutional layers using fast Fourier convolution and convolutional filter flow， taking into account the global contextual information of the image and local neighborhood information of the image pixel simultaneously. It learns to jointly optimize the measurement matrix and convolution filter flow and establishes a complete end-to-end trainable deep image reconstruction network. Verification experiments were performed on the Set5， Set11， and BSD68 test datasets commonly used in the field of compressive sensing image reconstruction at a 20% sampling rate. The image reconstruction quality of G2LNet was compared with that of the traditional algorithm MH and algorithm based on deep learning； the average peak signal-to-noise ratio of CSNet increased by 2.29 dB and 0.51 dB， respectively， effectively improving the quality of the reconstructed image..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2135 (2023)
Fine 3D reconstruction methods with Gaofen-7 satellite stereo images
Danchao GONG, Songlin LIU, Yilong HAN, and Wei ZHANG
To achieve high-precision and fine 3D reconstruction with the Gaofen-7 satellite sub-meter-level images， this paper proposes a method that focuses on the relative error correction of stereo pairs， image horizontal plane correction， and semi-global matching optimization， forming a fine 3D reconstruction pipeline. First，To achieve high-precision and fine 3D reconstruction with the Gaofen-7 satellite sub-meter-level images， this paper proposes a method that focuses on the relative error correction of stereo pairs， image horizontal plane correction， and semi-global matching optimization， forming a fine 3D reconstruction pipeline. First， concerning the relative error in the orientation model of Gaofen-7 stereo images， the geometric constraint relationship of the connection point among images is used to eliminate the systematic error of the rational function model. Second， a horizontal correction method based on the projection plane of the object is used to correct the original image； this eliminates the large inclination error difference between stereo images and provides a better data basis for subsequent processes. In the dense matching stage， global publicly available digital elevation model （DEM） data are used as the disparity constraint. AD-Census， which consider both the grayscale and feature information of the image， is employed as the matching cost metric， addressing the matching error problem caused by repeated texture and improving the digital surface model （DSM） production. The results of experiments conducted using Gaofen-7stereo images covering areas in Ningxia and Xinjiang indicate that the proposed method can improve the relative error accuracy from 0.847 and 0.725 pixels to 0.652 and 0.593 pixels， respectively， representing up to 23.02% improvement. The horizontal correction method based on the projection plane of the object can significantly eliminate the geometric distortion caused by the difference in large inclination angle， and good-quality DSM products can be obtained， especially for the repetitive texture of small-scale dense building areas..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2147 (2023)
Micro/Nano Technology and Fine Mechanics
CAD stereo model multi-source matching for ICF target pose measurement
ZhaoJie CHEN, Yang ZHOU, and Wei SONG
This study proposes a CAD stereo model matching-based positional measurement method to solve the high-precision positioning problem of tiny targets in high-power laser devices. The proposed method uses CAD file information to construct a stereo model of the target and correlate it with 3-CCD multi-source images for matThis study proposes a CAD stereo model matching-based positional measurement method to solve the high-precision positioning problem of tiny targets in high-power laser devices. The proposed method uses CAD file information to construct a stereo model of the target and correlate it with 3-CCD multi-source images for matching， avoiding the error superposition problem arising from the separate processing of multi-source images in traditional positioning methods. In this study， we defined the matching degree function of the target based on the Chamfer distance conversion of the model exploration point， which was used to evaluate the matching degree of the model and target. A genetic algorithm （GA） was used to find the optimal solution of the function to obtain the measured poses of the target. To address the problem of the GA being susceptible to a local optimum， a hierarchical adaptive genetic algorithm （HAGA） was developed to improve the search accuracy and efficiency of the algorithm. To improve the system debugging efficiency of complex engineering， this study used the OpenGL graphics library to build a simulation system to realize the functions of target simulation imaging and offline algorithm debugging. The simulation system avoids the interference of various factors， such as installation error， processing error， and camera ambient light in the system， and can debug and verify the algorithm conveniently as well as improve the engineering efficiency. An experiment was performed using a 3-CCD positioning measurement platform for the positioning adjustment of different target collimation states. The measurement results showed that the target translation and positioning error was less than 3.5 μm， and the angle error was less than 0.07°， which is less than the error requirement of the actual measurement system. This verifies the feasibility of the measurement algorithm in actual engineering applications..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2040 (2023)
Preparation of Al nanoarray on quartz and fluorescence enhancement performance
Ziheng WANG, Jinhua LI, Xueying CHU, and Ye ZHANG
To address the difficulties in the preparation of ultraviolet （UV） resonance metal nanoarrays on transparent substrates， Al nanoarrays were prepared by electron beam exposure on a non-conducting quartz substrate and their morphology and properties were studied. A Cr metal layer was introduced to overcome the problem ofTo address the difficulties in the preparation of ultraviolet （UV） resonance metal nanoarrays on transparent substrates， Al nanoarrays were prepared by electron beam exposure on a non-conducting quartz substrate and their morphology and properties were studied. A Cr metal layer was introduced to overcome the problem of non-conductivity in the electron beam exposure process. Nanorod morphology was optimized by adjustments to the exposure dose and design parameters. The electric field distribution of Al nanorods （excited by 325 nm UV light） was analyzed by using the finite-difference time-domain method. CdSe/ZnS quantum dots were deposited on the surface of Al nanoarrays by using Langmuir–Blodgett technology for fluorescence performance enhancement. The results indicate uniformity in the size distribution of Al nanorods prepared at a 1 200 μC/cm2 dose. Four regions with Al nanorods in mutually perpendicular orientations were integrated. Optical analysis results indicate that the fluorescence intensity of CdSe/ZnS quantum dots could be enhanced approximately 1.7 times by the Al nanorod array. Simulation results show that， under UV radiation， the electric field intensities at the ends of the nanorods were higher than at their sides， thereby revealing their influence on the fluorescence enhancement of the quantum dots. The multi-oriented integrated Al nanoarray with quantum dot fluorescence enhancement effects， which was prepared on a non-conducting substrate， provides a new strategy for the luminescence and polarization modulation of quantum dots under backlight excitation..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2052 (2023)
Surface crack depth detection of sapphire substrate two-sided lapping
Fuxin PENG, Zhongwei HU, Yu CHEN, Binhui XIE, and Zhihao ZHOU
Two-sided lapping is an essential process in the fabrication of sapphire substrates， and it significantly affects the amount of material removed during subsequent polishing. Therefore， studying the characteristics and measuring the depths of surface cracks on lapped substrates are important processes. In this study， weTwo-sided lapping is an essential process in the fabrication of sapphire substrates， and it significantly affects the amount of material removed during subsequent polishing. Therefore， studying the characteristics and measuring the depths of surface cracks on lapped substrates are important processes. In this study， we investigate the surface crack characteristics and crack depths of two-sided lapped sapphire substrates using section apparent micrometry， focused ion beam side observations， a differential etching rate method， a magnetorheological polishing method， and a layer-by-layer polishing method. Consequently， we observe subsurface cracks on the sapphire substrate after grinding using the cross-sectional microscopic observation and focused ion beam side observation methods. These cracks mainly include oblique lines， horizontal lines， hooks， and dendritic patterns. The differential etching rate method reveals that the thickness of the crack dense layer on the grinding surface of the sapphire substrate measures 9-10 μm. Using the magnetorheological polishing method， we measure the depth of local subsurface cracks on the grinding substrate to be 25-30 μm. Furthermore， employing the layer-by-layer polishing method， we determine that the overall subsurface crack depth of the ground substrate is approximately 30-35 μm. Additionally， based on the crack characteristics and depths detected using different methods， we construct a surface crack model for the two-sided lapping of sapphire substrates. This model serves as a foundation for formulating and optimizing subsequent polishing processes..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2060 (2023)
Research on post-processing polishing technology of pulse compression gratings with high laser damage threshold
Xiaolan HONG, and Chen JIANG
Pulsed compression gratings are critical optical components for the development of high-energy lasers. However， laser manufacturing processes often generate surface contaminants and microstructure defects， leading to technical challenges limiting the advancement of high-power laser systems. To improve the laser-inducedPulsed compression gratings are critical optical components for the development of high-energy lasers. However， laser manufacturing processes often generate surface contaminants and microstructure defects， leading to technical challenges limiting the advancement of high-power laser systems. To improve the laser-induced damage thresholds of gratings， a novel method involving magnetic compound fluid polishing for pulse compression gratings was developed herein. The microscopic structure， surface morphology， surface roughness， diffraction efficiency， and laser-induced damage thresholds of grating samples were evaluated before and after polishing. This assessment allowed a comparison of the grating surface quality and performance before and after polishing. Consequently， the magnetic compound fluid polishing process was found to effectively minimize burrs and microstructure defects generated during the manufacturing process without damaging the intrinsic grating structure. After 3 min of polishing， the grating surface roughness decreased from 21.36 nm to 3.73 nm. Furthermore， the laser damage threshold increased from 2.8 J/cm2 to 3.8 J/cm2， improving the laser damage resistance by 35.7% without influencing the diffraction efficiency. These results demonstrate that magnetic compound fluid polishing is a highly effective method for enhancing the surface quality and overall performance of grating components..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2071 (2023)
Modern Applied Optics
Design of auto-focusing system combining imaging characteristics of follow star system
Cao YAO, Wenyue ZHU, Wenqing XU, Gang XU, and Xiaowei CHEN
In severe defocused cases during stellar facula imaging， traditional gradient evaluation algorithms often present weak gradient details and fail to correctly identify the direction of focus. Here， we designed an automatic focusing system for the evaluation of star imaging characteristics to realize the rapid imaging anIn severe defocused cases during stellar facula imaging， traditional gradient evaluation algorithms often present weak gradient details and fail to correctly identify the direction of focus. Here， we designed an automatic focusing system for the evaluation of star imaging characteristics to realize the rapid imaging and reliable tracking of stars based on a star-following system； moreover， we studied the focusing evaluation algorithm， automatic focusing structure， and automatic control of the system. Differences in the focus and focus gradient of a facula image were analyzed based on the grayscale distribution of the star system images. The background value was included in the evaluation contribution using the characteristics of grayscale differences in stellar facula imaging， and the grayscale difference contribution of the facula signal was added to the traditional gradient evaluation function. To develop the automatic focusing system， three steps were followed： design of the focusing hardware and system structure adapted to the improved algorithm， construction of the focus assembly using a stepper motor and focus base， and the use of a calculation control unit to drive the focus assembly. Experiments and analyses were conducted using a star-tracking device equipped with the focusing system. The experimental results revealed that the autofocus device could focus steadily on the star-tracking system. The improved algorithm presented better focus orientation than the traditional gradient algorithm. The local extreme point was reduced to zero while maintaining the 67% sensitivity of the traditional gradient algorithm. Thus， the improved algorithm was not affected by weak gradient details. It could maintain better focus directionality and focus sensitivity at different defocus levels. Equipped with an autofocusing ability， the developed star-tracking system can rapidly determine a suitable focal distance and assess atmospheric optical characteristics..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2009 (2023)
Optical design of the off-axis reflective system with wide fov
Yaqiong YU, Lingjie WANG, Shangnan ZHAO, Jianping ZHANG, and Xin ZHANG
Generally， the observation ranges and detection efficiencies of optical remote sensors can be enhanced using wide-field optical imaging systems. Accordingly， the design and manufacturing of freeform-surface-based optical systems have achieved significant progress recently， making the construction of wide-field， rapid， Generally， the observation ranges and detection efficiencies of optical remote sensors can be enhanced using wide-field optical imaging systems. Accordingly， the design and manufacturing of freeform-surface-based optical systems have achieved significant progress recently， making the construction of wide-field， rapid， higher-quality-resolution and -imaging， and obstacle-free off-axis reflection systems feasible. The first part of this paper presents a detailed analysis of the aberration characteristics of an off-axis reflection system. This asymmetric higher level aberration increases dramatically as the field of view （FOV）， specifically the meridional FOV， of the system increases. These characteristics place a prominent requirement on image field continuity. The second part of this paper presents a new design approach for two-dimensional （2-D） large FOV optical systems with long focal lengths. First， we built multiple structural forms based on conventional optical systems， leading to a discretized meridional field angle， and with certain specific surfaces of the optical system decomposed into two sub-surfaces. Following this， system optimization was achieved by constraining the system size and improving its structural form. Finally， we designed the off-axis reflective freeform surface optical system. The imaging quality was good within the entire FOV， realizing 2-D large-field imaging of 40°×16°， a focal length of 1 m， and an F-number of 10. The test results prove that the proposed design is good， yielding an optical modulation transfer function of better than 0.26， particularly in the visible spectrum of 400-750 nm with a characteristic frequency of 50 lp/mm..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2019 (2023)
Polarized laser scattering detection method for subsurface microcracks in quartz glass
Qixin LÜ, Fangyuan SHI, Qian BAI, Xiaoxiong JIANG, and Lili JIANG
Subsurface microcracking is an unavoidable consequence of the quartz glass grinding process， and it significantly affects the service performance of optical components. Therefore， nondestructive detection of subsurface microcracks in quartz glass is crucial for grinding process optimization. Accordingly， herein， a polaSubsurface microcracking is an unavoidable consequence of the quartz glass grinding process， and it significantly affects the service performance of optical components. Therefore， nondestructive detection of subsurface microcracks in quartz glass is crucial for grinding process optimization. Accordingly， herein， a polarized laser scattering （PLS）-based method for detecting subsurface microcracks in quartz glass was developed. For this， a PLS nondestructive inspection platform was established. Indentation specimens with subsurface microcrack depths of 5.27， 9.7， and 15.42 μm were prepared via indentation experiments at pressure values of 20， 50， and 100 mN， respectively. Furthermore， ground specimens with subsurface microcrack depths ranging from 1-10 μm were prepared via grinding with different abrasive particle sizes （1-20 μm）. A power function relationship between the PLS detection signal and subsurface crack depth was discovered. The developed PLS detection system effectively detected and quantified subsurface microcracks with depths less than 10 μm. Thus， the developed PLS detection system enables microcrack detection in ground quartz glass， providing valuable guidance for subsurface microcrack control and process optimization..
Optics and Precision Engineering
- Publication Date: Jul. 25, 2023
- Vol. 31, Issue 14, 2031 (2023)
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20