检索结果-内蒙古大学图书馆

Journal of Network Intelligence 2022年第1期7卷 175-188页

作者： Zhao, Li-Quan Zhang, Yu-Peng Teng, Zi-Ming Jiang, Zi-Cong Cui, Ying Kan, Zhong-Feng Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology Ministry of Education Northeast Electric Power University Jilin132012 China College of information and communication Jilin University Changchun130012 China Graduate school of Information Science and Electrical Engineering Kyushu University Fukuoka819-0395 Japan Zhuhai Power Supply Bureau Guangdong Electric Power Corporation Zhuhai519000 China Jilin Power Supply Bureau Jilin Electric Power Corporation Jilin132012 China

Transfer learning is a deep learning method. It transforms knowledge in the source domain and feeds it to the target domain to improve the performance of the target domain model. The conditional adversarial domain adaption method is one of a number of transfer learning methods. It uses Resnet as the feature extraction network;however, Resnet can only determine the relationship between local features, and so its feature extraction accuracy is limited. To overcome this problem, a self-attention layer has been designed in order to determine the relationship among all features. This layer is added between the fourth layer and fifth layer of Resnet. Besides, a cross-entropy function is introduced into loss function, and a sigmoid layer is designed and used as the out layer in conditional adversarial domain adaption. These produce gradient disappearance in the deep network, and influence the discriminator training effect. To solve that problem, a least squares function is used as to design the loss function. The new type of loss function does not require the sigmoid function. Using theoretical analysis, we explain how this new loss function can be used to solve the gradient disappearance problem, thus rendering the network structure more stable. We compare our proposed method with other datasets, including Joint Adaptation Networks, adversarial domain adaptation, conditional adversarial domain adaption, Resnet-50, domain-adversarial neural network on imageCLEF-DA, Office-Home, and Office-31. Our simulation results show that the proposed conditional adversarial domain adaption based on self-attention method has the fastest convergence speed and highest accuracy the among different datasets. © 2022, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

MAE-GAN: A Novel Strategy for Simultaneous Super-resolution Reconstruction and Denoising of Post-stack Seismic Profile

arXiv

引用

arXiv 2024年

作者： Yu, Wenshuo Dong, Shiqi Lu, Shaoping Dong, Xintong The College of Instrumentation and Electrical Engineering Jilin University Jilin Changchun130026 China The Key Laboratory of Modern Power System Simulation and the Control and Renewable Energy Technology Ministry of Education Jilin Jilin City132012 China Department of Communication Engineering College of Electric Engineering Northeast Electric Power University Jilin Jilin City132012 China School of Earth Sciences and Engineering Sun Yat-Sen University Guangdong Guangzhou510275 China Guangdong Zhuhai519000 China Guangdong Provincial Key Lab of Geodynamics and Geohazards Sun Yat-Sen University Guangdong Guangzhou510275 China

Post-stack seismic profiles are images reflecting containing geological structures which provides a critical foundation for understanding the distribution of oil and gas resources. However, due to the limitations of seismic acquisition equipment and data collecting geometry, the post-stack profiles suffer from low resolution and strong noise issues, which severely affects subsequent seismic interpretation. To better enhance the spatial resolution and signal-to-noise ratio of post-seismic profiles, a multi-scale attention encoder-decoder network based on generative adversarial network (MAE-GAN) is proposed. This method improves the resolution of post-stack profiles, and effectively suppresses noises and recovers weak signals as well. A multi-scale residual module is proposed to extract geological features under different receptive fields. At the same time, an attention module is designed to further guide the network to focus on important feature information. Additionally, to better recover the global and local information of post-stack profiles, an adversarial network based on a Markov discriminator is proposed. Finally, by introducing an edge information preservation loss function, the conventional loss function of the Generative Adversarial Network is improved, which enables better recovery of the edge information of the original post-stack profiles. Experimental results on simulated and field post-stack profiles demonstrate that the proposed MAE-GAN method outperforms two advanced convolutional neural network-based methods in noise suppression and weak signal recovery. Furthermore, the profiles reconstructed by the MAE-GAN method preserve more geological structures. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Modeling and Simulation Research of a Novel Artificial Skin Sensor

引用

Journal of Physics: Conference Series 2020年第1期1693卷

作者： Feilu Wang Yong Hu Yang Song Yufeng Chen Ruigen Ye School of Electronics and Information Engineering Anhui Jianzhu University Hefei 230601 China Key laboratory of building information acquisition and measurement control technology Anhui Jianzhu University Hefei 230601 China

In this paper, a novel artificial skin sensor model with special structure based on the flexibility of Polydimethylsiloxane (PDMS) and the piezoresistive effect of Poly Vinylidene Fluoride (PVDF) piezoelectric fiber is proposed to solve the problems of structure design and complicated manufacturing process of skin sensor sensitive units. It has excellent piezoelectric effect and can detect the external force efficiently and precisely. The finite element analysis technology is used to simulate the deformation of the novel artificial skin sensor under external force from different aspects. Firstly, the crossing-mesh sensitive unit model and the three-layer configuration with three-layer of the artificial skin sensor are designed. Secondly, the external forces are applied to the sensitive units with different diameters by finite element analysis, and the sensitive unit with optimum diameter is figured out for the novel artificial skin sensor. After that, with the optimum sensitive unit, the performance of the artificial skin sensor model with the optimum sensitive unit is specifically simulated when it's subjected to different stress. Finally, through seriously analyzing the experiment results, the mapping relationship between the deformation of the artificial skin sensor and the external force is obtained. Experiment results demonstrate that the novel artificial skin sensor has high sensitivity, not only can truly reflect the approximate linear relationship between sensor deformation and external load, but also can accurately detect the magnitude of external force applied to the sensor.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Entropy feature extraction on flow pattern of gas/liquid two-phase flow based on cross-section measurement

引用

Journal of Physics: Conference Series 2009年第1期147卷

作者： J Han F Dong Y Y Xu Tianjin Key Laboratory of Process Measurement and Control School of Electrical Engineering & Automation Tianjin University Tianjin 300072 CHINA Department of Educational Technology Capital Normal University Beijing 100037 CHINA

This paper introduces the fundamental of cross-section measurement system based on Electrical Resistance Tomography (ERT). The measured data of four flow regimes of the gas/liquid two-phase flow in horizontal pipe flow are obtained by an ERT system. For the measured data, five entropies are extracted to analyze the experimental data according to the different flow regimes, and the analysis method is examined and compared in three different perspectives. The results indicate that three different perspectives of entropy-based feature extraction are sensitive to the flow pattern transition in gas/liquid two-phase flow. By analyzing the results of three different perspectives with the changes of gas/liquid two-phase flow parameters, the dynamic structures of gas/liquid two-phase flow is obtained, and they also provide an efficient supplementary to reveal the flow pattern transition mechanism of gas/liquid two-phase flow. Comparison of the three different methods of feature extraction shows that the appropriate entropy should be used for the identification and prediction of flow regimes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional force simulation prediction of flexible sensor based on BP neural network

引用

Journal of Physics: Conference Series 2020年第1期1678卷

作者： Feilu Wang Yufeng Chen Yang Song Yong Hu Rungen Ye Yanan Jiang School of Electronics and Information Engineering Anhui Jianzhu University Hefei 230601 China Key laboratory of building information acquisition and measurement control technology Anhui Jianzhu University Hefei 230601 China

When the robot grabs the object, the force information detection of the object is the basis for the smooth grabbing process. The force information of the object can be fully reflected by detecting the force in the three-dimensional direction. In this paper, with polydimethylsiloxane (PDMS) as the substrate and embedded into the sensitive unit prepared by conductive rubber, a new flexible sensor which can detect three-dimensional force is designed. Firstly, based on the piezoresistive effect of conductive rubber, COMSOL Multiphysics software was used to carry out multi-physical field simulation experiment for the sensor. Furthermore, based on the nonlinear approximation ability of BP neural network and the resistance of simulation output, the training sample set and the test sample set were constructed by using the 5-fold cross validation method, and the BP neural network model was constructed to achieve the accurate prediction of three-dimensional force. Finally, the number of hidden layer neurons was adjusted to optimize the BP network model. The results of cross-validation experiments show that the sensor designed in this paper can effectively detect the three-dimensional force information, and the optimized BP neural network can significantly improve the accuracy of the three-dimensional force prediction.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Domain Adaptation on Person Re-Identification via Dual-level Asymmetric Mutual Learning

arXiv

引用

arXiv 2023年

作者： Wu, Qiong Li, Jiahan Dai, Pingyang Ye, Qixiang Cao, Liujuan Wu, Yongjian Ji, Rongrong Institute of Artificial Intelligence The Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics Xiamen University Xiamen361005 China School of Information and Control Engineering China University of Mining and Technology Xuzhou221000 China The Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics Xiamen University Xiamen361005 China The Peng Cheng Laboratory Shenzhen518066 China The School of Electronics Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing100049 China The Media Analytics and Computing Lab Department of Computer Science School of Informatics Xiamen University Xiamen361005 China The Youtu Laboratory Tencent Shanghai200233 China The Fujian Engineering Research Center of Trusted Artificial Intelligence Analysis and Application Institute of Artificial Intelligence Xiamen University Xiamen361005 China

Unsupervised domain adaptation person re-identification (Re-ID) aims to identify pedestrian images within an unlabeled target domain with an auxiliary labeled source-domain dataset. Many existing works attempt to recover reliable identity information by considering multiple homogeneous networks. And take these generated labels to train the model in the target domain. However, these homogeneous networks identify people in approximate subspaces and equally exchange their knowledge with others or their mean net to improve their ability, inevitably limiting the scope of available knowledge and putting them into the same mistake. This paper proposes a Dual-level Asymmetric Mutual Learning method (DAML) to learn discriminative representations from a broader knowledge scope with diverse embedding spaces. Specifically, two heterogeneous networks mutually learn knowledge from asymmetric subspaces through the pseudo label generation in a hard distillation manner. The knowledge transfer between two networks is based on an asymmetric mutual learning manner. The teacher network learns to identify both the target and source domain while adapting to the target domain distribution based on the knowledge of the student. Meanwhile, the student network is trained on the target dataset and employs the ground-truth label through the knowledge of the teacher. Extensive experiments in Market-1501, CUHK-SYSU, and MSMT17 public datasets verified the superiority of DAML over state-of-the-arts. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Origins of Diamond Surface Noise Probed by Correlating Single-Spin measurements with Surface Spectroscopy

引用

Physical Review X 2019年第3期9卷 031052-031052页

作者： Sorawis Sangtawesin Bo L. Dwyer Srikanth Srinivasan James J. Allred Lila V. H. Rodgers Kristiaan De Greve Alastair Stacey Nikolai Dontschuk Kane M. O’Donnell Di Hu D. Andrew Evans Cherno Jaye Daniel A. Fischer Matthew L. Markham Daniel J. Twitchen Hongkun Park Mikhail D. Lukin Nathalie P. de Leon Department of Electrical Engineering Princeton University Princeton New Jersey 08544 USA Department of Physics Harvard University Cambridge Massachusetts 02138 USA Centre for Quantum Computation and Communication Technology School of Physics University of Melbourne Parkville Victoria 3010 Australia Department of Physics Astronomy and Medical Radiation Science Curtin University Bentley Western Australia 6102 Australia Department of Physics Aberystwyth University SY23 3BZ Aberystwyth United Kingdom Material Measurement Laboratory National Institute of Standards and Technology Gaithersburg Maryland 20899 USA Element Six Harwell OX11 0QR United Kingdom Department of Chemistry and Chemical Biology Harvard University Cambridge Massachusetts 02138 USA

The nitrogen-vacancy (NV) center in diamond exhibits spin-dependent fluorescence and long spin coherence times under ambient conditions, enabling applications in quantum information processing and sensing. NV centers near the surface can have strong interactions with external materials and spins, enabling new forms of nanoscale spectroscopy. However, NV spin coherence degrades within 100 nm of the surface, suggesting that diamond surfaces are plagued with ubiquitous defects. Prior work on characterizing near-surface noise has primarily relied on using NV centers themselves as probes; while this has the advantage of exquisite sensitivity, it provides only indirect information about the origin of the noise. Here we demonstrate that surface spectroscopy methods and single-spin measurements can be used as complementary diagnostics to understand sources of noise. We find that surface morphology is crucial for realizing reproducible chemical termination, and use this insight to achieve a highly ordered, oxygen-terminated surface with suppressed noise. We observe NV centers within 10 nm of the surface with coherence times extended by an order of magnitude.

关键词： NV centers Quantum sensing Surface & interfacial phenomena

来源：评论

学校读者我要写书评

暂无评论

Research on the Influence of Non-Conductor on the Weight Function of Electromagnetic Flowmeter

引用

Journal of Physics: Conference Series 2021年第3期1952卷

作者： Xuejing Li Lijun Sun Shanghai Institute of Measurement and Testing Technology Shanghai 201203 China Shanghai Key Laboratory of On-line Testing and Control Technology Shanghai 201203 China School of Electrical and Information Engineering Tianjin University Tianjin 300072 China.

The demand for multiphase flow measurement is widespread in petrochemical and other fields. Due to its small size, easy maintenance, and no moving parts, electromagnetic flowmeters are widely used in single fluid measurement. With the increasing application of traditional electromagnetic flowmeters in petrochemical and other fields, the problems of traditional electromagnetic flowmeters in multiphase flow measurement applications have been discovered. How to optimize the structure of the electromagnetic flowmeter and how to improve the measurement accuracy has become a new research focus. On the basis of a large number of literature studies, this research is based on the finite element method, and systematically studies the influence of multiphase flow on the weight function of electromagnetic flowmeters. Firstly, modeling the existence of non-conductive objects in the electromagnetic flowmeter. Secondly, analyze the weight function characteristics of the non-conductive objects on the electromagnetic flowmeter from two angles of different radii and different positions. The conclusions of this study can provide a certain reference basis for the electromagnetic flowmeter in the measurement of multiphase flow.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Investigating Correlation of Rough Surface Diffuse Scattering in Frequency Domain

Investigating Correlation of Rough Surface Diffuse Scatterin...

引用

European Conference on Antennas and Propagation, EuCAP

作者： Yang MIAO Mingming Gan Claude Oestges Katsuyuki Haneda Ke Guan Andrés Alayón Glazunov Wei Wang Yi Gong Dept. of Electrical and Electronic Engineering Southern University of Science and Technology Shenzhen China Universite catholique de Louvain Louvain-la-Neuve BE Institute of Information and Communication Technologies Electronics and Applied Mathematics (ICTM) Universit Catholique de Louvain Louvain-la-Neuve Belgium Dept. of Electronics and Nanoengineering Aalto University Espoo Finland State Key Lab of Rail Traffic Control and Safety Beijing Jiaotong University Beijing China Dept. of Electrical Engineering University of Twente Enschede Netherlands School of Information Engineering Chang’an University Xi’an China

This paper investigates numerically the correlation function in the frequency domain due to diffuse scattering originated from rough surfaces. The scattered fields are computed by means of physical optics methods. The irregularities of the rough surface are modeled as a correlated Gaussian process. The correlation of the channel transfer function comprising only diffuse scattering components is analyzed. In the analysis, radio propagation channels comprising orthogonally oriented dipoles at the link ends are considered while assuming different Gaussian roughness profiles. This study offers a useful characterization of the channel behavior in a multipath-rich environment due to diffuse scattering. The presented results are of particular interest to wireless systems based on orthogonal frequency-division multiplexing.

关键词： Rough surfaces Surface roughness Correlation Optical surface waves Scattering Surface impedance Frequency-domain analysis

来源：评论

学校读者我要写书评

暂无评论

[email protected] nanorods：Synthesis and their wide controllable range for the local surface plasmon

引用

Optical Materials 2022年 134卷

作者： Wenlong Yang Xingchen Meng Xinmei Liu Liuyang Zhang Chenwei Wang Xiaoyang Yu Ying Yang School of Measurement and Communication Engineering Harbin University of Science and Technology Harbin 150080 PR China Key Laboratory of Laser Spectroscopy Technology and Application of Heilongjiang Province Harbin 150080 PR China State Key Laboratory of Mechanics and Control of Mechanical Structures Nanjing University of Aeronautics and Astronautics Nanjing 210016 PR China School of Material Science and Chemical Engineering Harbin University of Science and Technology Harbin 150080 PR China

This work demonstrated the wide controllable range of the Local Surface Plasmon Resonance (LSPR) characteristic peak of the KTa x-1 Nb x O 3 @Ag ( [email protected] ). [email protected] nanocubes ( [email protected] NCs) and [email protected] nanorods ( [email protected] NRs) with different Ag loading rate have been prepared. Compared with [email protected] NCs, the LSPR characteristic peak of [email protected] NRs showed a wider controllable range and the enhanced factor could reach 5.48. This result could be derived from the higher aspect ratio of [email protected] NRs, resulting in an enhanced the coupling effect between KTN and Ag. This viewpoint could also be verified by the FDTD (Finite-Difference Time-Domain) calculation. With the increase of Ag loading rate, the coupling between KTN and Ag was weakened. And the characteristic peak of [email protected] show a blue shift, which shifts to the characteristic peak of Ag nanocrystal gradually. With the increase of Ag loading rate from 11 to 21 wt%, the LPSR characteristic peak of [email protected] NCs showed a blue shift from 569 to 547 nm, whereas peak for [email protected] NRs was shifted from 762 to 646 nm. This work combined the coupling effect of nanostructured KTN and Ag nanocrystals. And the anisotropy would be beneficial to broaden the LSPR controllable range of the characteristic extinction spectra of [email protected] Considering the low price of KTN, the [email protected] NRs should be a promising material applied in optical detection devices.

关键词： FDTD calculation LSPR Coupling effect Loading rate Blue shift

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：