检索结果-内蒙古大学图书馆

SPIE 12th International Symposium on Multispectral Image Processing and pattern recognition, MIPPR 2023

作者： Chen, Yi Zhu, Ying School of Electrical and Information Engineering Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan Institute of Technology Wuhan430205 China

ISBN: (数字)9781510675001

ISBN: (纸本)9781510674998

Approximately 50% of the Earth's surface is covered by clouds. optical remote sensing satellites encounter challenges in capturing clear ground images due to the difficulty of visible photoelectric magnetic waves penetrating clouds. Therefore, cloud detection is an essential and basic step in the processing and application of optical satellite remote sensing images. The traditional threshold-based cloud detection methods utilize spectral information of images to set thresholds, which is sensitive and less adaptable to varying imaging conditions. To enhance cloud detection performance, this paper proposes a deep learning method that combines residual network module and pyramid structure of dilated convolution. The method employs an encoder-decoder structure where the residual convolution module replaces conventional convolution kernels, reducing parameter quantity while enhancing feature expression capabilities. Furthermore, a pyramid structure of dilated convolution is introduced between the encoder and decoder to improve the acquisition of global information and reduce misjudgment of cloud pixels. In this work, ablation experiments are conducted to validate the reliability of the proposed network model. To evaluate the effectiveness of the proposed method, GaoFen-1 remote sensing image data was used for experimental verification. The results indicated that compared to traditional methods, the proposed approach achieves satisfactory cloud detection results in various surface types, including barren land, forests, grasslands/crops, and wetlands, while having a smaller model size and parameter quantity. © 2024 SPIE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Centerline Extraction of Line Structured Light in 3D Reconstruction 7

Centerline Extraction of Line Structured Light in 3D Reconst...

引用

7th International Conference on Electrical, Mechanical and Computer Engineering, ICEMCE 2023

作者： Zhang, Ji Wuhan Institute of Technology Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan China

ISBN: (纸本)9798350382877

At present, 3D reconstruction of binocular vision is a hot topic in the field of computer vision, and the extraction of laser fringe centers in line structured light images is a crucial step. In order to obtain the feature point set between the images, it is necessary to extract the centerline of the linear structured light bands. Firstly, through image preprocessing, perform image stereo correction and image Gaussian filtering, and adaptive threshold segmentation algorithm is used to segment the stripe area of the line structured light, and the image thinning algorithm is used to extract the initial centerline of the line structured light. Finally, the point set is constructed and the initial centerline is optimized twice by the gray-scale centroid method, and the centerline of the linear structured light stripe is accurately extracted. The results show that, compared with the Steger algorithm and the skeleton thinning method, the proposed method extracts the centerlines of the linear structured light stripes with higher accuracy. © 2023 IEEE.

关键词： Binocular vision

来源：评论

学校读者我要写书评

暂无评论

Compact topological edge modes through hybrid coupling of orbital angular momentum modes

引用

Physical Review A 2024年第6期109卷 063516-063516页

作者： Yang Wang Chuang Jiang Miaomiao Zhao Dong Zhao Shaolin Ke School of Electrical and Information Engineering Hubei Key Laboratory of Optical Information and Pattern Recognition

Topological waveguide arrays, which support robust light propagation through edge modes, offer a promising solution for fault-tolerant photonic chips. However, these arrays commonly require a considerable number of waveguide elements to ensure their topological protection, leading to increased footprint and fabrication costs. Here, we propose a topological waveguide array with only a few cells based on a photonic Aharonov-Bohm (AB) cage. This design is advantageous because the edge modes become compactly localized at the boundary waveguides. We achieve this by utilizing the orbital hybridization of the fundamental mode and first-order orbital angular momentum (OAM) modes to construct the AB cage, where arbitrary flux of artificial gauge fields (AGFs) can be generated and controlled by adjusting the angle between adjacent waveguides. Notably, these edge modes exhibit robustness against disorders of on-site potential and coupling, even within a small lattice. Furthermore, we demonstrate the extension of this mechanism to high-order OAM modes. Our work paves the way for miniaturizing topological devices using the orbital degree of freedom, holding the potential for exploring other AGF-enriched phenomena.

关键词： Topological effects in photonic systems Waveguides

来源：评论

学校读者我要写书评

暂无评论

Multifunctional light-field modulation based on hybrid nonlinear metasurfaces

引用

Chinese Physics B 2023年第10期32卷 73-89页

作者：钱树航王凯杨加兴关超龙华陆培祥 Wuhan National Laboratory for Optoelectronics and School of Physics Huazhong University of Science and TechnologyWuhan 430074China School of Electronic and Information Engineering Hubei University of Science and TechnologyXianning 437100China Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan Institute of TechnologyWuhan 430205China

The generation characteristics of nonlinear optical signals and their multi-dimensional modulation at micro-nano scale have become a prominent research area in nanophotonics,and also the key to developing various novel nonlinear photonics *** recent years,the demand for higher nonlinear conversion efficiency and device integration has led to the rapid progress of hybrid nonlinear metasurfaces composed of nanostructures and nonlinear *** a joint platform of stable wavefront modulation,nonlinear metasurface and efficient frequency conversion,hybrid nonlinear metasurfaces offer a splendid opportunity for developing the next-generation of multipurpose flat-optics *** article provides a comprehensive review of recent advances in hybrid nonlinear metasurfaces for light-field *** advantages of hybrid systems are discussed from the perspectives of multifunctional light-field modulation,valleytronic modulation,and quantum ***,the remaining challenges of hybrid metasurfaces are summarized and future developments are also prospected.

关键词： light-field modulation hybrid metasurfaces nonlinear optics two-dimensional materials

来源：评论

学校读者我要写书评

暂无评论

CTDPGAN: Infrared and Visible Image Fusion Using CNN-Transformer Dual-Process-Based Generative Adversarial Network 42

CTDPGAN: Infrared and Visible Image Fusion Using CNN-Transfo...

引用

42nd Chinese Control Conference, CCC 2023

作者： Zhang, Cheng Li, Xuan Xu, Yuhang Wang, Jie Chen, Rongfu Hubei Key Laboratory of Optical Information and Pattern Recognition China School of Electrical and Information Engineering Wuhan Institute of Technology China

ISBN: (纸本)9789887581543

Most of the existing image fusion methods prefer to use the adversarial learning game to fuse infrared and visible imagess. However, such single adversarial mechanism makes image fusion task easily ignore global contextual information. To this end, this paper proposes a CNN-Transformer dual-process-based generative adversarial network (CTDPGAN) to fuse infrared and visible images. In generator, a dual-process-based module composed by a CNN block and a Swin-Transformer block is proposed. The channel filter and spatial filter in the CNN block has the ability to adaptively extract additional complementary information from images of various modalities while preserving the shallow features of the source images. The Swin-Transformer Block (STRB) is designed to establish local attention by dividing non-overlapping windows and then to bridge global attention by interacting windows. In addition, we introduce generative adversarial learning networks into the training process, the dual-channel transformer discriminators are designed to improve the discriminative ability of the fused image. Thus, the fused image learns the distribution of global contextual information from source images and retain competitive visible light and infrared domains in more balanced manner. Moreover, we introduce the primary and auxiliary feature concepts into the structural similarity loss function and spatial frequency loss function, which will enable the generator to produce a fused image that retains thermal radiation information and rich detail information. Finally, the experimental findings demonstrate that, in both subjective and objective assessments, our model produces outcomes that are equivalent to or superior compared to state-of-the-art image fusion methods. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

Synthesis and Raman Performance Enhancement of Multilayer AuAg Heterostructures with Magnetic Resonance

引用

Wuhan University Journal of Natural Sciences 2023年第4期28卷 351-358页

作者： HUANG Fan ZHAO Zhirui FENG Jingru MA Liang DING Sijing School of Mathematics and Physics China University of Geosciences(Wuhan)Wuhan 430074HubeiChina Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan Institute of TechnologyWuhan 430205HubeiChina

Significant amplification of surface enhanced Raman scattering(SERS)signals can be achieved mainly by the electric field enhancement in metal core-shell nanostructures,and the enhanced magnetic field is rarely *** this study,we prepared multi-gap Au/AgAu core-shell hybrid nanostructures by using gold nanocup as the *** overgrowth processes to grow one,two,and three layers of AgAu hybrid nanoshells can produce Au/AgAu^(1),Au/AgAu^(2),and Au/AgAu^(3) *** strong plasmon coupling between the core and shell leads to significant electromagnetic field *** the synergistic effect of electromagnetic plasmon resonance and plasmon coupling,Au/AgAu core-shell hybrid nanostructures exhibit excellent SERS *** also investigate the effect of the interstitial position of the rhodamine B(RhB)molecule on Raman enhancement in Au/AgAu~3 *** study can provide new ideas for the synthesis of multi-gap Raman signal amplifiers based on magnetic plasmon coupling.

关键词： Au/AgAu core-shell hybrid nanostructures magnetic plasmon resonance plasmon coupling surface enhanced Raman scattering(SERS)

来源：评论

学校读者我要写书评

暂无评论

Epitaxial growth of 2D gallium selenide flakes for strong nonlinear optical response and visible-light photodetection

引用

Frontiers of physics 2023年第5期18卷 237-244页

作者： Mengting Song Nan An Yuke Zou Yue Zhang Wenjuan Huang Huayi Hou Xiangbai Chen Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan Institute of TechnologyWuhan 430205China State Key Laboratory of Materials Processing and Die&Mould Technology School of Materials Science and EngineeringHuazhong University of Science and TechnologyWuhan 430074China

As an emerging groupⅢ–Ⅵsemiconductor two-dimensional(2D)material,gallium selenide(GaSe)has attracted much attention due to its excellent optical and electrical *** this work,high-quality epitaxial growth of few-layer GaSe nanoflakes with different thickness is achieved via chemical vapor deposition(CVD)*** to the non-centrosymmetric structure,the grown GaSe nanoflakes exhibits excellent second harmonic generation(SHG).In addition,the constructed GaSe nanoflake-based photodetector exhibits stable and fast response under visible light excitation,with a rise time of 6 ms and decay time of 10 *** achievements clearly demonstrate the possibility of using GaSe nanoflake in the applications of nonlinear optics and(opto)-electronics.

关键词： 2D materials gallium selenide second harmonic generation chemical vapor deposition photodetector

来源：评论

学校读者我要写书评

暂无评论

TCPNet: A 3D Point Cloud Classification Model Based on the Advanced CNN and Transformer 3

TCPNet: A 3D Point Cloud Classification Model Based on the A...

引用

3rd International Conference on Artificial Intelligence and Computer information Technology, AICIT 2024

作者： Wu, Jinmeng Ma, Jingwei Hong, Hanyu Hao, Yanbin Ma, Lei Wang, Lei Wuhan Institute of Technology Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan China School of Information Science and Technology University of Science and Technology of China Anhui China

ISBN: (纸本)9798350365443

Transformer and Convolutional Neural Network (CNN) are currently two important models in the field of deep learning. Among them, Transformer has strong global perception ability but weak local perception ability, and CNN precisely compensates for this. CNN has strong local perception ability, so in order to combine the advantages of both, we propose a 3D classification model TCPNet based on Transformer and CNN. We first use Transformer to extract the global features of the point cloud, and then concatenate a CNN network to enhance the detail perception ability, so that the network can better understand the 3D point cloud. The experimental results on the ModelNet40 and ScanObjectNN datasets show that our method can effectively achieve 3D classification tasks. On ModelNet40, an accuracy of 93.6% can be achieved. On the most difficult variant PB-T50-RS of ScanObjectNN, an accuracy of 83.93% can be achieved. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Attention-guided Unsupervised Semantic Fusion for Infrared and Visible Images 15

Attention-guided Unsupervised Semantic Fusion for Infrared a...

引用

15th International Conference on Graphics and Image Processing, ICGIP 2023

作者： Qi, Qian Kong, Quan Liao, Qing Hu, Ziming Li, Jinglei Zhang, Jialian Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan Institute of Technology University China School of Art & Design Wuhan Institute of Technology University China

ISBN: (纸本)9781510675018

Infrared and visible image fusion aims to synthesize a new image with complementary information of the source images such as the thermal radiation information and detailed texture information. However, the existing methods applied in the image fusion tasks tend to pay more attention to the improvements of visual quality and multiple evaluation indicators, while ignoring the importance of semantic information in high-level missions of image processing. In this paper, we present a novel approach to image fusion, where the synthesis of fused images is achieved through the utilization of semantic layouts generated via unsupervised learning. With the introduction of Attention Mechanism, the relationships between each pair of pixel points are gained to construct soft semantic layouts and further capture the global context information to make the same kind of semantics show the same fusion effect. By leveraging the semantic information, our method enables the automatic learning of fusion weights for two source images at the same spatial position. In comparison with other state-of-the-art image fusion methods, our experimental results obtained excellent performance both on qualitative results and quantitative indicators. Moreover, our method retains the high-level semantic information to the greatest extent, which is one of our outstanding characteristics. © 2024 SPIE. All rights reserved.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

Unveiling Nonsecular Collisional Dissipation of Molecular Alignment

引用

Physical Review Letters 2024年第3期133卷 033202-033202页

作者： Pu Wang Lixin He Yu Deng Siqi Sun Pengfei Lan Peixiang Lu and School of Physics Hubei Optical Fundamental Research Center Wuhan 430074 China Hubei Key Laboratory of Optical Information and Pattern Recognition

We conducted a joint theoretical and experimental study to investigate the collisional dissipation of molecular alignment. By comparing experimental measurements to the quantum simulations, the nonsecular effect in the collision dissipation of molecular alignment was unveiled from the gas-density-dependent decay rates of the molecular alignment revival signals. Different from the conventional perspective that the nonsecular collisional effect rapidly fades within the initial few picoseconds following laser excitation, our simulations of the time-dependent decoherence process demonstrated that this effect can last for tens of picoseconds in the low-pressure regime. This extended timescale allows for the distinct identification of the nonsecular effect from molecular alignment signals. Our findings present the pioneering evidence that nonsecular molecular collisional dissipation can endure over an extended temporal span, challenging established concepts and strengthening our understanding of molecular dynamics within dissipative environments.

关键词： Atomic & molecular processes in external fields Rotational states

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：