检索结果-内蒙古大学图书馆

Journal of Network Intelligence 2024年第2期9卷 1046-1055页

作者： Zhang, Fu-Quan Ma, Lin-Juan Zhou, Jun-Xian Wang, Run-Hong Zhang, Guang-Zhen College of Computer and Control Engineering Minjiang University Fuzhou China Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music Chengdu China Fuzhou Technology Innovation Center of Intelligent Manufacturing Information System Minjiang University Fuzhou China Fujian Province University Fuzhou China School of Computer Science and Technology Beijing Institute of Technology Beijing China Minjiang University Fuzhou China Information Media School of Tianjin Technician Institute of Mechanical & Electrical Technology Tianjin China

The purpose of this paper is to present a new GAN-ID algorithm to gener-ate interior scene intelligently based on Artificial Intelligence of Things (AIoT). Although great progress has been made when GAN algorithm applied in smart homes and smart cities, the issue of the trade-off between the accuracy and the efficiency in interior scene has been overlooked. A number of experiments were performed to check the performance and it is indicated that the results of the proposed GAN-ID has great advantages in model sample generation than other common methods when generating interior scene intelligently, which is similar to state-of-the-art. It is meaningful when it can be applied in smart cities and smart homes filed based on AIoT in the future. © 2024, J. Network Intell. All rights reserved.

关键词： Automation

来源：评论

学校读者我要写书评

暂无评论

Steel Surface Defect Detection Based on Improved YOLOv8 20

Steel Surface Defect Detection Based on Improved YOLOv8

引用

20th International Conference on Natural Computation, Fuzzy systems and Knowledge Discovery, ICNC-FSKD 2024

作者： Zhang, Jichen Zhang, Zhi College of Computer Science and Technology Wuhan University of Science and Technology Hubei Wuhan430065 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Hubei Wuhan430065 China Big Data Science and Engineering Research Institute Wuhan University of Science and Technology Hubei Wuhan430065 China

ISBN: (纸本)9798350356328

There are some defects on the surface of steel that are difficult to detect, so we propose an improved algorithm based on YOLOv8 for detecting defects that are difficult to be detected. This improvement includes the introduction of an SPD module to minimize the problem of information loss due to downsampling;the addition of a small object detection head combined with EMA attention to further enhance the detection capabilities for small targets;the use of BiFPN to replace the original PANet, applying different weights to different features to strengthen feature fusion. Experiments were conducted on the NEU-DET dataset, and the algorithm improved mAP to 80.2%, an improvement of 3.9% over the baseline model. © 2024 IEEE.

关键词： Change detection

来源：评论

学校读者我要写书评

暂无评论

FELSDiff: Image Super-Resolution Based on Diffusion Model Fusing Feature Extraction with Latent Space 20

FELSDiff: Image Super-Resolution Based on Diffusion Model Fu...

引用

20th International Conference on Natural Computation, Fuzzy systems and Knowledge Discovery, ICNC-FSKD 2024

作者： Wu, Cancan Zhang, Zhi Wuhan University of Science and Technology College of Computer Science and Technology Hubei Wuhan430065 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan430065 China Big Data Science and Engineering Research Institute Wuhan University of Science and Technology Hubei Wuhan430065 China

ISBN: (纸本)9798350356328

In addressing issues like an excessive number of network parameters and the loss of fine texture details in image super-resolution reconstruction approaches based on diffusion models, this passage introduces a novel image super-resolution model (FELSDiff). This model relies on the fusion of diffusion models, feature extraction, and latent space. FELSDiff utilizes a Pre-trained CNN for initial super-resolution to extract the primary low-band features and part high-band features of the image. The preliminary super-resolved image is then concatenated as a condition to this model. We introduce a cross-attention mechanism guided by the high-band features of the preliminary super-resolved image to enhance the network's feature extraction capability and obtain more effective features. In the U-Net structure, we incorporate a latent space strategy, enabling the model's training and inference processes to be conducted in the latent space, thereby reducing the number of network parameters. Experimental results on multiple standard datasets indicate that this method outperforms other advanced methods overall. © 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

Learning Knowledge Evolution with Time Duration from Finer-Time-Granularity Temporal Knowledge Graph

Learning Knowledge Evolution with Time Duration from Finer-T...

引用

Chinese Automation Congress (CAC)

作者： Shaochong Du Kang Chen Shijie Yang Hong Huo Tao Fang Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education Shanghai China Shanghai Engineering Research Centre of Intelligent Control and Management Shanghai China

How to represent the temporal information corresponding to each fact in temporal knowledge graphs (TKGs) effectively is always challenging. Most existing representation learning methods usually map the timelines of knowledges into multiple points and ignore the underlying continuous time property and periodic characteristics behind those dense timestamps in finer-time granularity temporal knowledge graphs. A novel Knowledge Evolution with Time Duration (KETD) model is proposed for knowledge representation of temporal knowledge graphs with finer-time granularities. It represents the entity embedding as a time-varying nonlinear function to find the specific continuous time property from dense timestamps. To the best of our knowledge, it is the first attempt to learn embeddings of time durations and combines them into embeddings of entities and relations to predict potential periodic facts. It also uses the temporal point process to capture the impact of historical facts on the current fact. The experiments on self-built finer-time granularity TKG of Satellite-to-Earth Communication (STEC) and two public datasets have demonstrated the superiority of KETD compared to some baseline approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Forest Fire Recognition Method Based on R-shufflenetv2 5

A Real-Time Forest Fire Recognition Method Based on R-shuffl...

引用

5th International Symposium on Autonomous systems, ISAS 2022

作者： Li, Mengna Zhang, Youmin Mu, Lingxia Xin, Jing Xue, Xianghong Jiao, Shangbin Liu, Han Xie, Guo Yi, Yingmin Xi'an University of Technology Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Shaanxi Xi'an710048 China Concordia University Department of Mechanical Industrial and Aerospace Engineering MontrealQCH3G 1M8 Canada

ISBN: (纸本)9781665487085

The recognition of early forest fires can reduce the resource loss caused by fire combustion. A real-time forest fire image recognition method based on r-shufflenetv2 network is proposed. R-shufflenetv2 is mainly composed of a series of r-shufflenetv2 units and is an improved version of the shufflenetv2 network. In order to improve the recognition accuracy, the shufflenetv2 unit is reconstructed by using the residual structure, which enhances the feature extraction ability of the network. The experimental results show that the r-shufflenetv2 network is a good fire recognition model. On the benchmark FLAME dataset, r-shufflenetv2 has higher fire recognition accuracy than the original shufflenetv2. In addition, r-shufflenetv2 also achieves real-time detection, and its recognition speed is about 31 FPS. © 2022 IEEE.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

FG-UAP: Feature-Gathering Universal Adversarial Perturbation

arXiv

引用

arXiv 2022年

作者： Ye, Zhixing Cheng, Xinwen Huang, Xiaolin Department of Automation The MOE Key Laboratory of System Control and Information Processing Shanghai Jiao Tong University Shanghai200240 China

Deep Neural Networks (DNNs) are susceptible to elaborately designed perturbations, whether such perturbations are dependent or independent of images. The latter one, called Universal Adversarial Perturbation (UAP), is very attractive for model robustness analysis, since its independence of input reveals the intrinsic characteristics of the model. Relatively, another interesting observation is Neural Collapse (NC), which means the feature variability may collapse during the terminal phase of training. Motivated by this, we propose to generate UAP by attacking the layer where NC phenomenon happens. Because of NC, the proposed attack could gather all the natural images’ features to its surrounding, which is hence called Feature-Gathering UAP (FG-UAP). We evaluate the effectiveness our proposed algorithm on abundant experiments, including untargeted and targeted universal attacks, attacks under limited dataset, and transfer-based black-box attacks among different architectures including Vision Transformers, which are believed to be more robust. Furthermore, we investigate FG-UAP in the view of NC by analyzing the labels and extracted features of adversarial examples, finding that collapse phenomenon becomes stronger after the model is corrupted. The code will be released when the paper is accepted. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Adaptive Cohen's Class Time-Frequency Distribution

arXiv

引用

arXiv 2024年

作者： Cui, Manjun Zhang, Zhichao Han, Jie Chen, Yunjie Cao, Chunzheng The School of Mathematics and Statistics Nanjing University of Information Science and Technology Nanjing210044 China The Key Laboratory of System Control and Information Processing Ministry of Education Shanghai200240 China The Key Laboratory of Computational Science and Application of Hainan Province Haikou571158 China The School of Remote Sensing and Geomatics Engineering Nanjing University of Information Science and Technology Nanjing210044 China

The fixed kernel function-based Cohen's class time-frequency distributions (CCTFDs) allow flexibility in denoising for some specific polluted signals. Due to the limitation of fixed kernel functions, however, from the view point of filtering they fail to automatically adjust the response according to the change of signal to adapt to different signal characteristics. In this letter, we integrate Wiener filter principle and the time-frequency filtering mechanism of CCTFD to design the least-squares adaptive filter method in the Wigner-Ville distribution (WVD) domain, giving birth to the least-squares adaptive filter-based CCTFD whose kernel function can be adjusted with the input signal automatically to achieve the minimum mean-square error denoising in the WVD domain. Some examples are also carried out to demonstrate that the proposed adaptive CCTFD outperforms some state-of-the-arts in noise suppression. Copyright © 2024, The Authors. All rights reserved.

关键词： Adaptive filters

来源：评论

学校读者我要写书评

暂无评论

Chaos synchronization of semiconductor lasers over 1040-km fiber relay transmission with hybrid amplification

引用

Photonics Research 2023年第6期11卷 953-960页

作者： LONGSHENG WANG JUNLI WANG YUSHAN WU YUEHUI SUN SONGSUI LI LIANSHAN YAN YUNCAI WANG ANBANG WANG Key Laboratory of Advanced Transducers and Intelligent Control System Ministry of EducationTaiyuan 030024China College of Physics and Optoelectronics Taiyuan University of TechnologyTaiyuan 030024China School of Information Engineering Guangdong University of TechnologyGuangzhou 510006China Guangdong Provincial Key Laboratory of Photonics Information Technology Guangzhou 510006China Center for Information Photonics&Communications School of Information Science and TechnologySouthwest Jiaotong UniversityChengdu 610031China

Optical chaos communication and key distribution have been extensively demonstrated with high-speed advantage but only within the metropolitan-area network range of which the transmission distance is restricted to around 300 *** secure-transmission requirement of the backbone fiber link,the critical threshold is to realize long-reach chaos ***,we propose and demonstrate a scheme of long-reach chaos synchronization using fiber relay transmission with hybrid amplification of an erbium-doped fiber amplifier(EDFA)and a distributed fiber Raman amplifier(DFRA).Experiments and simulations show that the hybrid amplification extends the chaos-fidelity transmission distance thanks to that the low-noise DFRA suppresses the amplified spontaneous emission noise and self-phase *** of the hybrid-relay conditions are studied,including launching power,gain ratio of DFRA to EDFA,single-span fiber length,and number of fiber span.A 1040-km chaos synchronization with a synchronization coefficient beyond 0.90 is experimentally achieved,which underlies the backbone network-oriented optical chaos communication and key distribution.

关键词： synchronization fiber amplifier

来源：评论

学校读者我要写书评

暂无评论

Retraction Note: A new sequence optimization algorithm based on particle swarm for machine learning (Journal of Ambient Intelligence and Humanized Computing, (2022), 13, 5, (2601-2619), 10.1007/s12652-021-03004-3)

引用

Journal of Ambient Intelligence and Humanized Computing 2024年第1期15卷 295-295页

作者： Xie, Chaofan Zhang, Fuquan Electronic Information and Engineering Institute Fuqing Branch of Fujian Normal University Fuzhou China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Minjiang University Fuzhou China

https://***/10.1007/s12652-021-03004-3. The Publisher has retracted this article in agreement with the Editor-in-Chief. The article was submitted to be part of a guest-edited issue. An investigation by the publisher found a number of articles, including this one, with a number of concerns, including but not limited to compromised editorial handling and peer review process, inappropriate or irrelevant references or not being in scope of the journal or guest-edited issue. Based on the investigation’s findings the publisher, in consultation with the Editor-in-Chief therefore no longer has confidence in the results and conclusions of this article. The authors have not responded to correspondence regarding this retraction. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Chaotic Neural Network with Nonlinear Delayed Self-feedback and Applications

Chaotic Neural Network with Nonlinear Delayed Self-feedback ...

引用

2023 International Conference on Applied Mathematics, Modelling and Statistics Application, AMMSA 2023

作者： Zhang, Yu Xu, Liuyang Xu, Yaoqun Chen, Bin Liao, Qian School of Computer and Information Engineering Harbin University of Commerce Harbin150028 China Heilongjiang Provincial Key Laboratory of Electronic Commerce and Information Processing Harbin150028 China Heilongjiang Cultural Big Data Theory Application Research Center Harbin150028 China Institute of System Engineering Harbin University of Commerce Harbin150028 China

We propose a novel transient chaotic neural network model with a nonlinear delayed self-feedback term, aiming to leverage its complex dynamical behavior for algorithm optimization. To avoid the influence of chance, we conduct experiments using randomly generated data and analyze the model's chaotic dynamical behavior by visualizing the stability of the neurons through inverse bifurcation diagrams and maximum Lyapunov exponent diagrams. The simulation results demonstrate the remarkable effectiveness of this new model in network optimization, with a success rate even reaching 100%, surpassing previous chaotic neural network models. Notably, in the following sections, all nonlinear terms are represented using trigonometric functions. © 2023 Institute of Physics Publishing. All rights reserved.

关键词： Traveling salesman problem

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：