检索结果-内蒙古大学图书馆

IEEE/CVF International Conference on Computer Vision (ICCV)

作者： Quan, Yuhui Huang, Haoran He, Shengfeng Xu, Ruotao South China University of Technology School of Computer Science and Engineering Guangzhou 510006 China Pazhou Lab Guangzhou 510335 China Singapore Management University School of Computing and Information Systems 188065 Singapore

ISBN: (纸本)9798350307184

Removing moire patterns from videos recorded on screens or complex textures is known as video demoireing. It is a challenging task as both structures and textures of an image usually exhibit strong periodic patterns, which thus are easily confused with moire patterns and can be significantly erased in the removal process. By interpreting video demoireing as a multi-frame decomposition problem, we propose a compact invertible dyadic network called CIDNet that progressively decouples latent frames and the moire patterns from an input video sequence. Using a dyadic cross-scale coupling structure with coupling layers tailored for multi-scale processing, CIDNet aims at disentangling the features of image patterns from that of moire patterns at different scales, while retaining all latent image features to facilitate reconstruction. In addition, a compressed form for the networks output is introduced to reduce computational complexity and alleviate overfitting. The experiments show that CIDNet outperforms existing methods and enjoys the advantages in model size and computational efficiency.

关键词： Deep Video Computational Efficiency Image Features Model Size Video Sequences Image Pattern Latent Features Compressed Format Complex Texture Multiscale Process Latent Image Deep Neural Network Attention Mechanism Video Frames Peak Signal-to-noise Ratio Multiscale Analysis Feature Alignment Adjacent Frames Supplement For Details Discrete Cosine Transform Frame Alignment U-shaped Structure Alignment Blocks Multi-scale Structure Reversible Loss Color Shift information Fidelity Latent Code Spatial Pyramid Pooling Spatial Pyramid

来源：评论

学校读者我要写书评

暂无评论

Intelligent research and judgment model of fire threshold based on clustering analysis

Intelligent research and judgment model of fire threshold ba...

引用

2023 International Conference on Networking, Informatics and Computing, ICNETIC 2023

作者： Meng, Xiankui Lin, Miaoling Liu, Jingang Guangdong Polytechnic Normal University School of Mathematics and Systems Science Guangzhou China Sun Yat-Sen University Guangdong Provincial Key Laboratory of Information Security Technology Guangzhou China

ISBN: (纸本)9798350313314

With the continuous progress of modern society and the rapid improvement of scientific and technological innovation ability, people's living standards are improving day by day. However, people enjoy a convenient life at the same time, the fire hazard is greatly increased. Once a fire occurs, it will bring serious harm. According to the statistics of the World Fire Statistics Center (WFSC) of the United Nations, there are about 7 million fires in the world every year, with about 65000 to 75000 deaths. In order to avoid fire and reduce the loss caused by fire, a good intelligent fire alarm system and a timely alarm mechanism are very necessary. Therefore, the reliability and failure rate of fire detectors become the key parameters that need to be balanced, in this paper, the actual number of fires in the city from June 1 to June 18 was determined to be 418 through the screening of MATLBA software, and the reliability and failure rate of each early-warning component were quantitatively analyzed. After that, CRITIC and Delphi methods were used to combine weight assignment. Through Cronbach's Alpha test, TOPSIS algorithm was improved. The evaluation coefficient was defined by Soergel and Kulczynskl. The improved TOPSIS-SK model was used to evaluate various types of components. According to the evaluation scores, the government was helped to select reliable fire detector types as point smoke detector and manual alarm button, the biggest innovation of this paper is to improve the TOPSIS comprehensive evaluation, and introduce the Zeger abundance and Kakshinski distance to simplify the calculation process;Secondly, the established intelligent research and judgment model of alarm threshold has strong universality and is of great significance to reality. © 2023 IEEE.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

A Short Review of FPGA-Accelerated Deep Learning for ECG Signal Classification in Precision Healthcare

A Short Review of FPGA-Accelerated Deep Learning for ECG Sig...

引用

World Conference on Complex systems (WCCS)

作者： Hanane El Ferdaoussi Wissam Jenkal Mostafa Laaboubi Bouchra Bendahane Rachid Latif Laboratory of Systems Engineering and Information Technology (LISTI) ENSA Ibn Zohr University Agadir Morocco

ISBN: (数字)9798331519131

ISBN: (纸本)9798331519148

Precision healthcare has advanced with the intro-duction of Field Programmable Gate Arrays for real-time ECG signal processing. The potential of FPGA-based acceleration of deep learning to increase diagnostic efficiency and accuracy for ECG signal classification is discussed. The paper summarizes current techniques and implementations based on state-of-the-art deep learning architectures, including Convolutional Neural Networks, which have shown notable improvements in computing efficiency and accuracy in classification when implemented on FPGA systems. In addition, a comparative analysis of exper-imental results is provided in the research so as to illustrate the generalizability and robustness of a variety of deep learning models. A comparative analysis of performance by categorization technologies shows the relative benefits of FPGA-based solutions regarding speed, accuracy, and resource efficiency.

关键词： Deep learning Accuracy Pattern classification Medical services Electrocardiography Real-time systems Sustainable development Field programmable gate arrays Optimization Monitoring

来源：评论

学校读者我要写书评

暂无评论

AI algorithms for Autonomous Robotics Navigation 2

AI algorithms for Autonomous Robotics Navigation

引用

2nd International Conference Computational and Characterization Techniques in engineering and sciences, IC3TES 2024

作者： Komala, C.R. Raja, J. Marotrao, Sarange Shreepad Neelima, Nalla Sujatha, R. Chidambarathanu, K. East Point College of Engineering and Technology Department of Information Science and Engineering Karnataka Bengaluru560049 India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering School of Computing Tamil Nadu Chennai600062 India Ajeenkya D Y Patil School of Engineering Department of Mechanical Engineering Maharashtra Pune412105 India R.V.R & J.C. College of Engineering Department of Information Technology Andhra Pradesh Chowdavaram522019 India Subbalakshmi Lakshmipathy College of Science Department of Computer Science Tamil Nadu Madurai625022 India R.M.K. Engineering College Department of Computer Science and Business Systems Tamil Nadu Chennai601206 India

ISBN: (纸本)9798350364699

Robotic systems depend much on autonomous navigation since it allows them to operate on their own in environments constantly changing and lacking clarity. The most evolved artificial intelligence (AI) techniques applied in self-driving robot navigation are briefly reviewed in this work. These approaches call for classical methodologies, machine learning, and deep learning as well. We examine the advantages and drawbacks of numerous approaches applied in route planning, location finding, map creation, obstacle avoidance. We investigate how neural networks, sensor fusion, and reinforcement learning may be applied to provide increased dependability and flexibility for guiding systems. Case examples illustrating the advantages and drawbacks of applying artificial intelligence (AI) in practical contexts including service robots, drones, and self-driving cars abound. Combining modern AI algorithms with conventional approaches indicates that it is feasible to create self-driving robots more dependable and efficient based on the increased tracking performance. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Modelling of all optical Synchronous Up-counter using microring resonator 9

Modelling of all optical Synchronous Up-counter using micror...

引用

Nanophotonics and Micro/Nano Optics IX 2023

作者： Sharma, Anjali Choure, Kamal Kishor Saharia, Ankur Mudgal, Nitesh Tiwari, Manish Singh, Ghanshyam Kumar, Santosh Department of Electronics and Communication Engineering Malaviya National Institute of Technology Jaipur India Manipal University Jaipur India Poornima College of Engineering Jaipur India Shandong Key Laboratory of Optical Communication Science and Technology School of Physics Science and Information Technology Liaocheng University Liaocheng China

ISBN: (纸本)9781510667952

The current work proposes an optimized design for 2-bit synchronous up counter using silicon nitride on insulator optical microring resonators (MRRs). The research involves exploring the existing designs of up counter structures based on MRRs, and then minimizing the number of rings being used in the proposed model. The proposed design uses four MRRs for Synchronous up counter operation and the model has been validated through the MATLAB simulation result. The main component of the design is the MRR which can act as a switch depending upon the output at the drop port or the through port according to the MRR being in an on-resonance or off-resonance state respectively. The principle of the MRR is that on applying a vertical pump signal over the ring, the non-linear refractive index of the ring undergoes a temporary blue shift resulting due to π- phase shift in the ring, changing the resonant frequency of operation of the ring. The average pump power required for the blue shift can be obtained from the phase shift vs. the average pump power graph simulated in MATLAB. This shift in resonance frequency is employed to switching action for the signal to be output at the drop port or the through port of the MRR and based on the drive signal inputs of the present state and the clock to the rings in the resonator structure which act as pump signal over the ring, the next state of the counter is determined. © 2023 SPIE. All rights reserved.

关键词： Silicon nitride

来源：评论

学校读者我要写书评

暂无评论

Adaptive Occlusion Face Recognition Feature Fusion Network Based on Precise Guidance Mask Extraction

引用

Journal of Shanghai Jiaotong University (science) 2025年 1-12页

作者： Ye, Jihua Zhang, Lei Wang, Chao Tu, Qixuan Zheng, Rehua Huang, Huiyuan Jiang, Aiwen School of Computer and Information Engineering Jiangxi Normal University Nanchang330022 China Jiangxi Provincial Key Laboratory for High Performance Computing Jiangxi Normal University Nanchang330022 China State International Science & Technology Cooperation Base of Networked Supporting Software Jiangxi Normal University Nanchang330022 China

The uncertainty in the position and size of occluding objects greatly affects the extraction of identity features in facial recognition, which is a challenge that existing methods fail to effectively address. To tackle this issue, this paper introduces a novel feature fusion network for occluded facial recognition that adaptively handles arbitrary occlusions and accurately extracts occlusion information masks. The network effectively mitigates the impact of occlusions on facial identity features through a feature fusion module. The proposed method employs an occlusion classification module and an occlusion classification loss to precisely guide the extraction of accurate occlusion masks. Subsequently, an occlusion-aware hybrid attention module is introduced, which focuses on the relationship between facial features and occlusion masks across channel and spatial dimensions. This fusion strategy enables the extraction of facial identity features that are resilient to occlusions. Experimental results indicate that the proposed method holds a significant advantage in handling facial recognition tasks with arbitrary occlusions. On the LFW and MF1 synthetic occlusion datasets, the average accuracy rates reached 93.19% and 72.57%, respectively, showing improvements of 1.7 and 2.39 percent points over mainstream methods. © Shanghai Jiao Tong University 2025.

关键词： A feature fusion occluded facial recognition occlusion mask occlusion perception TP391.41

来源：评论

学校读者我要写书评

暂无评论

Source-configured symmetry-broken hyperbolic polaritons

引用

eLight 2023年第1期3卷 136-146页

作者： Caixing Hu Tian Sun Ying Zeng Weiliang Ma Zhigao Dai Xiaosheng Yang Xinliang Zhang Peining Li Wuhan National Laboratory for Optoelectronics and School of Optical and Electronic Information Huazhong University of Science and TechnologyWuhan 430074China Optics Valley Laboratory Hubei 430074China Faculty of Materials Science and Chemistry Engineering Research Center of NanoGeomaterials of Ministry of EducationChina University of GeosciencesWuhan 430074China Xidian University Xi’an 710126China

Polaritons are quasi-particles that combine light with matter,enabling precise control of light at deep subwavelength *** excitation and propagation of polaritons are closely linked to the structural symmetries of the host materials,resulting in symmetrical polariton propagation in high-symmetry ***,in low-symmetry crystals,symmetry-broken polaritons exist,exhibiting enhanced directionality of polariton propagation for nanoscale light manipulation and ***,we theoretically propose and experimentally demonstrate the existence of symmetry-broken polaritons,with hyperbolic dispersion,in a high-symmetry *** show that an optical disk-antenna positioned on the crystal surface can act as an in-plane polarized excitation source,enabling dynamic tailoring of the asymmetry of hyperbolic polariton propagation in the high-symmetry crystal over a broad frequency ***,we provide an intuitive analysis model that predicts the condition under which the asymmetric polaritonic behavior is maximized,which is corroborated by our simulations and *** results demonstrate that the directionality of polariton propagation can be conveniently configured,independent of the structure symmetry of crystals,providing a tuning knob for the polaritonic response and in-plane anisotropy in nanophotonic applications.

关键词： Hyperbolic polaritons Near-field excitation Crystal symmetry In-plane polarization Calcite

来源：评论

学校读者我要写书评

暂无评论

A Patch-Based Multiple Imputation Transformer Model for Time Series Data Completion

A Patch-Based Multiple Imputation Transformer Model for Time...

引用

International Conference on Big Data and information Analytics (BigDIA)

作者： Wenbo Yuan Chaofan Dai Huahua Ding Qideng Tang National Key Laboratory of Information Systems Engineering National University of Defense Technology ChangSha China

ISBN: (数字)9798350354621

ISBN: (纸本)9798350354638

In the real world, collecting time series data often encounters difficulties and uncertainties, leading to prevalent data missing issues. The time series imputation task aims to restore data integrity and improve prediction performance. Existing models mainly use single imputation to handle missing data. This paper proposes a Patch-based Multiple Imputation Transformer (PMIT), which learns data correlations within the same feature through patch operations and channel-independent (CI) operations. The model employs multiple imputation training, optimizing imputed data propagation across different modules, and incorporates attention mechanisms and dynamic adjustment methods to address complex dependencies. Experimental results show that PMIT outperforms recent models in imputation performance and generalization ability across various realworld datasets.

关键词： Training Analytical models Uncertainty Data integrity Time series analysis Predictive models Transformers Imputation Data models Robustness

来源：评论

学校读者我要写书评

暂无评论

Ultrabroadband chiral metasurface for linear polarization conversion and asymmetric transmission based on enhanced interference theory

引用

Chinese Optics Letters 2023年第11期21卷 147-154页

作者：赵京城李南程用志 School of Electronics Information Engineering Beihang UniversityBeijing 100191China Aerospace Institute of Advanced Material&Processing Technology Beijing 100074China School of Information Science and Engineering Wuhan University of Science and TechnologyWuhan 430081China Hubei Longzhong Laboratory Xiangyang 441000China

In this paper,we propose an ultrabroadband chiral metasurface(CMS)composed of S-shaped resonator structures situated between two twisted subwavelength gratings and dielectric *** innovative structure enables ultrabroadband and high-efficiency linear polarization(LP)conversion,as well as asymmetric transmission(AT)effect in the microwave *** enhanced interference effect of the Fabry-Perot-like resonance cavity greatly expands the bandwidth and efficiency of LP conversion and AT *** numerical simulations,it has been revealed that the cross-polarization transmission coefficients for normal forward(-z)and backward(+z)incidence exceed 0.8 in the frequency range of 4.13 to 17.34 GHz,accompanied by a polarization conversion ratio of over 99%.Furthermore,our microwave experimental results validate the consistency among simulation,theory,and ***,we elucidate the distinct characteristics of ultrabroadband LP conversion and significant AT effect through analysis of polarization azimuth rotation and ellipticity angles,total transmittance,AT coefficient,and electric field *** proposed CMS structure shows excellent polarization conversion properties via AT effect and has potential applications in areas such as radar,remote sensing,and satellite communication.

关键词： chiral metasurface linear polarization conversion asymmetric transmission Fabry-Perot-like resonance electromagnetic interference model

来源：评论

学校读者我要写书评

暂无评论

Modeling of Sensor Enabled IrrigationManagement for Intelligent Agriculture Using Hybrid Deep Belief Network

引用

Computer systems science & engineering 2023年第8期46卷 2319-2335页

作者： Saud Yonbawi Sultan Alahmari B.R.S.S.Raju Chukka Hari Govinda Rao Mohamad Khairi Ishak Hend Khalid Alkahtani JoséVarela-Aldás Samih M.Mostafa Department of Software Engineering College of Computer Science and EngineeringUniversity of JeddahJeddahSaudi Arabia King Abdul Aziz City for Science and Technology RiyadhKingdom of Saudi Arabia Aditya Engineering College Affiliated to JNTUK(Kakinada) SurampalemIndia Department of MBA Vignan’s Institute of Information and Technology(A)DuvvadaVisakhapatnamAP530049India School of Electrical and Electronic Engineering Engineering CampusUniversiti Sains Malaysia(USM)Nibong TebalPenang14300Malaysia Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11564Saudi Arabia SISAu Research Group Universidad Tecnológica IndoaméricaAmbato180103Ecuador Faculty of Computers and Information South Valley UniversityQena83523Egypt

Artificial intelligence(AI)technologies and sensors have recently received significant interest in intellectual *** the application of AI technologies and agriculture sensors in intellectual agriculture is urgently required for the growth of modern agriculture and will help promote smart *** irrigation scheduling systems were highly required in the agricultural field due to their capability to manage and save water deficit irrigation *** learning systems devise an alternative to conventional irrigation management through the automatic elaboration of predictions related to the learning of an *** this motivation,this study develops a modified black widow optimization with a deep belief network-based smart irrigation system(MBWODBN-SIS)for intelligent *** MBWODBN-SIS algorithm primarily enables the Internet of Things(IoT)based sensors to collect data forwarded to the cloud server for examination ***,the MBWODBN-SIS technique applies the deep belief network(DBN)model for different types of irrigation classification:average,high needed,highly not needed,and not *** MBWO algorithm is used for the hyperparameter tuning process.A wideranging experiment was conducted,and the comparison study stated the enhanced outcomes of the MBWODBN-SIS approach to other DL models with maximum accuracy of 95.73%.

关键词： Agriculture smart farming hyperparameter tuning artificial intelligence irrigation management sensors deep learning

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：