Journals > > Topics > Optics at Surfaces
Optics at Surfaces|74 Article(s)
Optimized Design of Terahertz Coding Frequency Selection Surface Based on Deep Learning
Pan Zhou, Lei Gong, Zhiqiang Yang, Liguo Wang, Lihong Yang, Yao Li, Haibin Wang, and Jie Yu
This study proposes an intelligent integration-design method to address the problems of complex structure, numerous parameters, and time-consuming optimization of terahertz frequency selective surface (FSS) cells. The method is based on convolutional neural network (CNN) combined with genetic algorithm, and is applied to the performance optimization design of typical filtered FSS. The FSS periodic cell topology is encoded as a 16×16"0/1"rotationally symmetric sequence, and 26,000 groups of transmission and reflectance spectra in the range of 0.5?3 THz are collected as the dataset. A 19-layer CNN is used to obtain the spectral prediction with an average absolute error as low as 0.06 on the test set. The difference between the predicted and the target spectra is used to give a generalized objective function for the design of various typical FSS cells. Combined with the genetic algorithm, a single-frequency bandpass and bandstop FSSs with a bandwidth of 0.1 THz, a single-frequency bandpass and bandstop FSSs with a bandwidth of 0.5 THz, and a dual-frequency bandpassFSS with a bandwidth of 0.2 THz are designed and implemented, with good polarization stability for all of them. Computational results show that various typical bandpass and bandstop FSS cells can be realized concisely and efficiently by optimizing their topological coding. This study proposes an intelligent integration-design method to address the problems of complex structure, numerous parameters, and time-consuming optimization of terahertz frequency selective surface (FSS) cells. The method is based on convolutional neural network (CNN) combined with genetic algorithm, and is applied to the performance optimization design of typical filtered FSS. The FSS periodic cell topology is encoded as a 16×16"0/1"rotationally symmetric sequence, and 26,000 groups of transmission and reflectance spectra in the range of 0.5?3 THz are collected as the dataset. A 19-layer CNN is used to obtain the spectral prediction with an average absolute error as low as 0.06 on the test set. The difference between the predicted and the target spectra is used to give a generalized objective function for the design of various typical FSS cells. Combined with the genetic algorithm, a single-frequency bandpass and bandstop FSSs with a bandwidth of 0.1 THz, a single-frequency bandpass and bandstop FSSs with a bandwidth of 0.5 THz, and a dual-frequency bandpassFSS with a bandwidth of 0.2 THz are designed and implemented, with good polarization stability for all of them. Computational results show that various typical bandpass and bandstop FSS cells can be realized concisely and efficiently by optimizing their topological coding.
Laser & Optoelectronics Progress
- Publication Date: Feb. 25, 2024
- Vol. 61, Issue 4, 0424001 (2024)
Plasmonic Tunnel Junctions (Invited)
Yuanjia Feng, Junsheng Zheng, Ruoxue Yang, and Pan Wang
Plasmonic tunnel junctions are metal-insulator-metal structures with a nanoscale dielectric gap that can simultaneously support surface plasmons and electron tunneling. The extremely strong confinements and interactions of electrons, plasmons, and photons in the junctions provide a novel platform, merging electronics and photonics for studying and manipulating electrons and photons at the nanoscale. In this review, the recent progress in the field of plasmonic tunnel junctions is overviewed as concerns their applications in plasmon/light generation and plasmon/light-electron conversion. Plasmonic tunnel junctions are metal-insulator-metal structures with a nanoscale dielectric gap that can simultaneously support surface plasmons and electron tunneling. The extremely strong confinements and interactions of electrons, plasmons, and photons in the junctions provide a novel platform, merging electronics and photonics for studying and manipulating electrons and photons at the nanoscale. In this review, the recent progress in the field of plasmonic tunnel junctions is overviewed as concerns their applications in plasmon/light generation and plasmon/light-electron conversion.
Laser & Optoelectronics Progress
- Publication Date: Feb. 10, 2024
- Vol. 61, Issue 3, 0324001 (2024)
Three-Dimensional Measurement Method of Mirror Based on Telecentric Lens and Curved Screen
Yu Li, Zonghua Zhang, Nan Gao, Zhaozong Meng, Ziyu Li, and Zhangying Wang
A new method based on a telecentric lens and curved screen is proposed to measure a discontinuous mirror's three-dimensional (3D) shape. This technique increases the imaging range of mirrors having large curvature and improves the measurement accuracy of 3D data. First, a display screen shows sinusoidal fringe patterns, and a camera records the sinusoidal fringe images reflected through the plane reference mirror and the measured mirror. Second, the corresponding phase distribution is obtained using the four-step phase shift method and the optimal three fringe selection method. The phase change modulated by the object surface of the measured mirror is obtained by comparison to the plane reference mirror. Furthermore, according to the established mathematical model, the relationship between phase and depth is deduced, and the system parameters are calibrated. Finally, measurements are taken to verify the accuracy and effectiveness of the proposed method using an artificial mirror step with large curvature and a discontinuous mirror. A new method based on a telecentric lens and curved screen is proposed to measure a discontinuous mirror's three-dimensional (3D) shape. This technique increases the imaging range of mirrors having large curvature and improves the measurement accuracy of 3D data. First, a display screen shows sinusoidal fringe patterns, and a camera records the sinusoidal fringe images reflected through the plane reference mirror and the measured mirror. Second, the corresponding phase distribution is obtained using the four-step phase shift method and the optimal three fringe selection method. The phase change modulated by the object surface of the measured mirror is obtained by comparison to the plane reference mirror. Furthermore, according to the established mathematical model, the relationship between phase and depth is deduced, and the system parameters are calibrated. Finally, measurements are taken to verify the accuracy and effectiveness of the proposed method using an artificial mirror step with large curvature and a discontinuous mirror.
Laser & Optoelectronics Progress
- Publication Date: Apr. 10, 2023
- Vol. 60, Issue 7, 0724001 (2023)
Metal Workpiece Surface Defect Segmentation Method Based on Improved U-Net
Yi Wang, Xiaojie Gong, and Jia Cheng
To solve the problem of low segmentation accuracy of metal workpiece surface defects, we propose a workpiece surface defect segmentation model based on a U-net network combined with a multi-scale adaptive-pattern feature extraction and bottleneck attention module. First, we embed a multi-feature attention aggregation module in the network to improve the utilization of information and extract more relevant features, so as to extract defect targets with high accuracy. Then, the bottleneck attention modules are introduced into the network to increase the weight of defect targets, optimize the extraction of features, and obtain more feature information, thus obtaining better segmentation accuracy. The improved network mean pixel accuracy reaches 0.8749, which is 2.92% higher than the original network. The mean intersection over union reaches 0.8625, an increase of 3.72%. Compared to the original network, the improved network has better segmentation accuracy and segmentation results. To solve the problem of low segmentation accuracy of metal workpiece surface defects, we propose a workpiece surface defect segmentation model based on a U-net network combined with a multi-scale adaptive-pattern feature extraction and bottleneck attention module. First, we embed a multi-feature attention aggregation module in the network to improve the utilization of information and extract more relevant features, so as to extract defect targets with high accuracy. Then, the bottleneck attention modules are introduced into the network to increase the weight of defect targets, optimize the extraction of features, and obtain more feature information, thus obtaining better segmentation accuracy. The improved network mean pixel accuracy reaches 0.8749, which is 2.92% higher than the original network. The mean intersection over union reaches 0.8625, an increase of 3.72%. Compared to the original network, the improved network has better segmentation accuracy and segmentation results.
Laser & Optoelectronics Progress
- Publication Date: Aug. 10, 2023
- Vol. 60, Issue 15, 1524001 (2023)
Study on Surface-Enhanced Infrared Absorption Characteristics of Double-Layer Petal-Structure Optical Antenna
Xingzheng Shi, Chun Li, Xiaoyan Fan, Guang Yuan, Wanmei Sun, Lin Xiao, Zhedong Wang, and Shaodong Wang
A double-layer petal-structure optical antenna for mid-infrared detection is presented in this paper. The finite-difference time-domain method is used to analyze the effects of structural parameters and polarization direction of incident light on the resonant wavelength of the antenna and intensity of the electric field at the tip of the antenna. Based on the optimization of a single-layer structure, the ratios of the intensity of the electric field at the tip of the upper antenna to the intensity of the incident light are calculated with different incident wavelengths when the gap (h) between the two antennas is 0.1-0.8 μm. To investigate the enhancing mechanism of the lower antenna on the electric field of the upper antenna, the variations of the electric-field intensity ratio of the same measured point with and without the upper antenna are analyzed under fixed incident wavelength and an enlarged gap h (0.1-3.6 μm). The results indicate that the enhancement of the electric field at the tip of the upper antenna is mainly attributed to the coupling effect of the double-layer antenna structure, with h being less than 1 μm. When h is less than 0.2 μm, the electric-field intensity at the tip of the upper antenna reduces with decreasing h. This is because the energy at the tip of upper antenna is transferred to the interlayer region owing to near-field coupling. However, when h exceeds 1 μm, the enhancement of the electric field at the tip of the upper antenna is mainly ascribed to the interference effect of the reflected light from the lower antenna. A double-layer petal-structure optical antenna for mid-infrared detection is presented in this paper. The finite-difference time-domain method is used to analyze the effects of structural parameters and polarization direction of incident light on the resonant wavelength of the antenna and intensity of the electric field at the tip of the antenna. Based on the optimization of a single-layer structure, the ratios of the intensity of the electric field at the tip of the upper antenna to the intensity of the incident light are calculated with different incident wavelengths when the gap (h) between the two antennas is 0.1-0.8 μm. To investigate the enhancing mechanism of the lower antenna on the electric field of the upper antenna, the variations of the electric-field intensity ratio of the same measured point with and without the upper antenna are analyzed under fixed incident wavelength and an enlarged gap h (0.1-3.6 μm). The results indicate that the enhancement of the electric field at the tip of the upper antenna is mainly attributed to the coupling effect of the double-layer antenna structure, with h being less than 1 μm. When h is less than 0.2 μm, the electric-field intensity at the tip of the upper antenna reduces with decreasing h. This is because the energy at the tip of upper antenna is transferred to the interlayer region owing to near-field coupling. However, when h exceeds 1 μm, the enhancement of the electric field at the tip of the upper antenna is mainly ascribed to the interference effect of the reflected light from the lower antenna.
Laser & Optoelectronics Progress
- Publication Date: Jan. 10, 2023
- Vol. 60, Issue 1, 0124001 (2023)
Process Optimization of CNTs/Cu Composite Coating Prepared by Laser-Assisted Low Pressure Cold Spraying
Chaowei Jiang, Bo Liu, Yunyi Zhang, Jingyong Sun, Bo Li, Qunli Zhang, and Jianhua Yao
In order to meet the needs of repair and surface modification of copper contacts in high voltage switchgears, carbon nanotubes (CNTs)/Cu composite coating is prepared on the surface of Cu substrate by laser-assisted low pressure cold spraying with CNTs as reinforcing phase and Cu as bonding phase. First, the surface of CNTs is metallized by electroless-plating of copper to improve the density of CNTs and the interfacial bonding between CNTs and Cu. Then, the spraying distance, scanning speed, and laser irradiation power are optimized. Finally, the microstructure of the composite coating is analyzed by energy dispersion spectroscopy (EDS) and scanning electron microscope (SEM). The results show that the interfacial bonding between the coating and the substrate is the best when the spraying distance is 15 mm. With the increase of scanning speed, the internal of the coating has relatively good compactness and coating/interface bonding state, and the surface of the coating is smooth. With the increase of laser irradiation power, the thickness and width of the coating first increase and then decrease, and the coating thickness reaches the maximum at 800 W. Due to the softening effect of laser heating and the protective effect of copper films on the surface of CNTs, CNTs can be evenly dispersed in the composite coating and maintain the structural integrity. In order to meet the needs of repair and surface modification of copper contacts in high voltage switchgears, carbon nanotubes (CNTs)/Cu composite coating is prepared on the surface of Cu substrate by laser-assisted low pressure cold spraying with CNTs as reinforcing phase and Cu as bonding phase. First, the surface of CNTs is metallized by electroless-plating of copper to improve the density of CNTs and the interfacial bonding between CNTs and Cu. Then, the spraying distance, scanning speed, and laser irradiation power are optimized. Finally, the microstructure of the composite coating is analyzed by energy dispersion spectroscopy (EDS) and scanning electron microscope (SEM). The results show that the interfacial bonding between the coating and the substrate is the best when the spraying distance is 15 mm. With the increase of scanning speed, the internal of the coating has relatively good compactness and coating/interface bonding state, and the surface of the coating is smooth. With the increase of laser irradiation power, the thickness and width of the coating first increase and then decrease, and the coating thickness reaches the maximum at 800 W. Due to the softening effect of laser heating and the protective effect of copper films on the surface of CNTs, CNTs can be evenly dispersed in the composite coating and maintain the structural integrity.
Laser & Optoelectronics Progress
- Publication Date: Apr. 10, 2022
- Vol. 59, Issue 7, 0724001 (2022)
Phase Distribution of Laguerre-Gaussian Beams on a Layered Topological Insulator Dielectric Film
Hui Meng, Mingjun Wang, Duo Ning, and Shenhe Ren
Herein, the phase distribution characteristics of Laguerre-Gaussian (LG) beams incident on a layered topological insulator (TI) thin film are studied using the plane angular spectrum expansion method and transmission matrix theory. Numerical results show that the phase structure of LG beams in the reflected and transmitted fields is affected by the changes in topological magneto-electric polarizability. Particularly, the center axis of the phase distribution of p-wave shifts left or right. This research on the phase distribution characteristics of LG beams incident on layered topological insulator thin films has significance in wireless laser communication, optical trapping, particle manipulation, nonlinear optics, information coding, and other fields. Herein, the phase distribution characteristics of Laguerre-Gaussian (LG) beams incident on a layered topological insulator (TI) thin film are studied using the plane angular spectrum expansion method and transmission matrix theory. Numerical results show that the phase structure of LG beams in the reflected and transmitted fields is affected by the changes in topological magneto-electric polarizability. Particularly, the center axis of the phase distribution of p-wave shifts left or right. This research on the phase distribution characteristics of LG beams incident on layered topological insulator thin films has significance in wireless laser communication, optical trapping, particle manipulation, nonlinear optics, information coding, and other fields.
Laser & Optoelectronics Progress
- Publication Date: Mar. 10, 2022
- Vol. 59, Issue 5, 0524002 (2022)
Simulation of Ag Nano Pit-Silicon Grating Composite Microstructures Used for with Near Infrared Absorption Enhancement
Guobin Sun, Jin Zhang, Shilei Jiang, and Liu Yang
Aiming at the characteristics of two-dimensional silicon grating microstructure in the near infrared light absorption rate is extremely low, a micro-nano composite structure is proposed to enhance the near infrared absorption of Ag nano-pit and silicon grating by using isolators. The absorption rate of the composite structure in the wavelength range from 0.78 μm to 2.5 μm is studied based on the finite-difference time-domain method. The influence of the grating structure and Ag nano-pits on the light absorption efficiency is analyzed. The simulation results show that when Ag nano pits with a diameter of 0.1 μm are embedded in the grating gap with a period of 0.2 μm and a duty cycle of 0.5 and on the surface of the grid column, the absorptivity of the composite microstructure is above 23% in the near infrared wide band, and the average absorptivity is up to 52.3%, theoretically. When Al2O3 dielectric layer is added on the grating surface, the absorption rate of wide band is above 41.3%, and the average absorption rate is increased to 65.1%. Finally, the absorption efficiency is improved in the wide wavelength range of near infrared, which provides a new method to enhance optical absorption in photodetector, solar cell, optical communication, radar stealth, biomedical and so on. Aiming at the characteristics of two-dimensional silicon grating microstructure in the near infrared light absorption rate is extremely low, a micro-nano composite structure is proposed to enhance the near infrared absorption of Ag nano-pit and silicon grating by using isolators. The absorption rate of the composite structure in the wavelength range from 0.78 μm to 2.5 μm is studied based on the finite-difference time-domain method. The influence of the grating structure and Ag nano-pits on the light absorption efficiency is analyzed. The simulation results show that when Ag nano pits with a diameter of 0.1 μm are embedded in the grating gap with a period of 0.2 μm and a duty cycle of 0.5 and on the surface of the grid column, the absorptivity of the composite microstructure is above 23% in the near infrared wide band, and the average absorptivity is up to 52.3%, theoretically. When Al2O3 dielectric layer is added on the grating surface, the absorption rate of wide band is above 41.3%, and the average absorption rate is increased to 65.1%. Finally, the absorption efficiency is improved in the wide wavelength range of near infrared, which provides a new method to enhance optical absorption in photodetector, solar cell, optical communication, radar stealth, biomedical and so on.
Laser & Optoelectronics Progress
- Publication Date: Mar. 10, 2022
- Vol. 59, Issue 5, 0524001 (2022)
Texture Characteristics of Polarized Thermal Images on Metal Surfaces in Fatigue Damage Process
Darong Zhu, Shanji Yang, Fangbin Wang, Jingfa Lei, Dagui Wang, and Qinglei Luan
Infrared thermography is an effective method for the detection of metal fatigue damage conditions. However, it does not consider the effect of surface microstructure on spontaneous emission in the fatigue damage process; thus, interpreting the infrared thermal image characteristics from a microcosmic perspective is difficult. Metal fatigue is a complex energy dissipation process, and spontaneous emission has polarization properties. By incorporating polarization detection into infrared thermography, besides temperature field information, surface texture information including emissivity change can also be obtained. Based on the foregoing, this paper obtained the thermal infrared polarization images of the metal material surface under cyclic load using Q235 low carbon steel as the research object by constructing a tensile fatigue test platform and a polarization thermal image acquisition platform. We investigated the evolution processes of surface morphology of metal materials during metal fatigue damage by using the gray level co-occurrence matrix (GLCM) to extract the thermal image texture information, such as polarization azimuth, polarization degree, and Stokes parameters. The results of the experiments show that the texture features of polarized thermal images of metal components change with fatigue damage cycles, and the co-occurrence matrix statistics exhibit some regularity with cycles. Infrared thermography is an effective method for the detection of metal fatigue damage conditions. However, it does not consider the effect of surface microstructure on spontaneous emission in the fatigue damage process; thus, interpreting the infrared thermal image characteristics from a microcosmic perspective is difficult. Metal fatigue is a complex energy dissipation process, and spontaneous emission has polarization properties. By incorporating polarization detection into infrared thermography, besides temperature field information, surface texture information including emissivity change can also be obtained. Based on the foregoing, this paper obtained the thermal infrared polarization images of the metal material surface under cyclic load using Q235 low carbon steel as the research object by constructing a tensile fatigue test platform and a polarization thermal image acquisition platform. We investigated the evolution processes of surface morphology of metal materials during metal fatigue damage by using the gray level co-occurrence matrix (GLCM) to extract the thermal image texture information, such as polarization azimuth, polarization degree, and Stokes parameters. The results of the experiments show that the texture features of polarized thermal images of metal components change with fatigue damage cycles, and the co-occurrence matrix statistics exhibit some regularity with cycles.
Laser & Optoelectronics Progress
- Publication Date: Feb. 20, 2022
- Vol. 59, Issue 4, 0424001 (2022)
Milling Surface Roughness Measurement Under Few-Shot Problem
Huaian Yi, Runji Fang, Aihua Shu, and Enhui Lu
Most machine vision-based roughness measurement methods either build a prediction model based on roughness correlation indices or build an index-free prediction model using deep learning networks. However, both these models have disadvantages. The artificial designed index has a complicated calculation process, which is not conducive to inline detection. In comparison, deep learning models rely heavily on big data. It is difficult to train an effective model when the amount of data is insufficient. To address the above problems, this study proposes a graph neural network-based method for measuring the roughness of milling surfaces. This proposed approach acquired the ability to learn autonomously during the training phase. Thus, only a few milling samples were required to measure the roughness of the milling workpieces. The experimental results show that the proposed method can automatically extract features on roughness measurement of milling workpieces with high accuracy and good robustness of lighting environment. Most machine vision-based roughness measurement methods either build a prediction model based on roughness correlation indices or build an index-free prediction model using deep learning networks. However, both these models have disadvantages. The artificial designed index has a complicated calculation process, which is not conducive to inline detection. In comparison, deep learning models rely heavily on big data. It is difficult to train an effective model when the amount of data is insufficient. To address the above problems, this study proposes a graph neural network-based method for measuring the roughness of milling surfaces. This proposed approach acquired the ability to learn autonomously during the training phase. Thus, only a few milling samples were required to measure the roughness of the milling workpieces. The experimental results show that the proposed method can automatically extract features on roughness measurement of milling workpieces with high accuracy and good robustness of lighting environment.
Laser & Optoelectronics Progress
- Publication Date: Dec. 10, 2022
- Vol. 59, Issue 23, 2324001 (2022)
Topics
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20