检索结果-内蒙古大学图书馆

A Text Detection Method Based on Multi-Scale Selective Fusion Feature Pyramid and Multi-Semantic Spatial Network for Visual IoT

引用

IEEE Internet of Things Journal 2025年第11期12卷 15997-16008页

作者： Wang, Manli Dou, Zeya Chen, Wei Wang, Zehua Yu, Fei Richard Leung, Victor C. M. Henan Polytechnic University School of Physics and Electronic Information Henan Jiaozuo454003 China China University of Mining and Technology School of Computer Science and Technology Jiangsu Xuzhou221116 China China University of Mining and Technology School of Mechanical Electronic and Information Engineering Beijing100083 China The University of British Columbia Department of Electrical and Computer Engineering VancouverBCV6T 1Z4 Canada Carleton University School of Information Technology OttawaONK1S 5B6 Canada

With the rapid development of Visual Internet of Things (VIoT) and text detection technology, they have been widely combined and applied to many industrial production sites, such as label text detection, achieving impressive results. However, there are still many shortcomings in the text detection technology: 1) The existing VIoT system has very limited detection precision for text with large scale changes, especially for some small-scale text detection;2) The existing text detection algorithms cannot meet the actual situation, as the labels often contain handwritten texts, and the text to be detected is arbitrary shape;3) In the actual detection, there are many creases or defects on the text label. To solve the above problems, this paper designs a text detection method based on a multi-scale selection fusion feature pyramid and multi-semantic spatial network to assist the VIoT system in detecting label text. Firstly, a multi-scale selective fusion feature pyramid is designed, which not only uses the texture extraction module to effectively improve the text texture feature and multi-scale feature extraction ability, but also uses the cross-scale selective fusion block to selectively fuse the features of different stages to reduce the influence of pollution on detection. In addition, a multi-semantic spatial network is designed to capture the multi-semantic spatial information of each feature channel by using the multi-scale deep shared one-dimensional convolution, which effectively integrates global context dependence and multi-semantic spatial prior. Experimental results show that the comprehensive index F-measure on the public datasets ICDAR2015, Total-Text and CTW1500 is increased by 5.7%, 3.3%and 3.8% respectively. Furthermore, the precision, recall, and F-measure on the dataset Label-Text are 94.6%, 90.7%and 92.6% respectively. The label text detection VIoT system we designed has been deployed in the field and achieved excellent performance. The code of our pro

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Data Protection in Enhanced EXplainable Digital Twins: Insights from Hybrid AI 2

Data Protection in Enhanced EXplainable Digital Twins: Insig...

引用

2nd International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2025

作者： Gupta, Prashant K. Mazumdar, Bireshwar Dass Pillai, Subash Nataraja School of Computer Science & Engineering Technology Bennett University Uttar Pradesh Greater Noida India Division of Agricultural Physics ICAR-Indian Agricultural Research Institute Pusa Campus New Delhi110012 India

ISBN: (纸本)9798331530389

Deep Learning (DL) based artificial intelligence (AI) is becoming pervasive in almost all the day to day applications. These DL algorithms require large amounts of data, which can be easily collected nowadays by deploying sensors in the environment of the system under observation. These DL algorithms however, suffer from drawbacks such as being criticised as 'black-boxes', being biased to generating predictions based on past data and ignoring the human expertise. Hence, an architecture of Enhanced eXplainable Digital Twins (EXDTs) was proposed. EXDTs remove 'black-boxness' of DL algorithms by using the XAI approaches. Further, the digital twins (DTs) underlying the EXDTs are constructed through the sensor data and depict the real-time simulatability. Also, to improve the explainability and use the human expertise/experience, EXDTs use the domain experts' qualitative data. EXDTs are a class of Hybrid AI systems, that use the numeric data for underlying DL algorithms and subjective knowledge of the domain expert. This multimodal data in EXDTs opens up new attacks fronts for data manipulation and change the EXDTs' outputs. In this manuscript, we discuss about the data protection in EXDTs, by simulating various types of attacks and suggesting possible ways the system can counter them. We feel our proposed framework is useful for the involved stakeholders in various domains like precision agriculture, healthcare, manufacturing, etc. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Predictions of H-mode access and edge pedestal instability in the EHL-2 spherical torus

引用

Plasma science and Technology 2025年第2期27卷 54-63页

作者： Yumin WANG Kai LI Zhuo HUANG Yiliang LIU Shuyu DAI Jie ZHANG Yanqing HUANG Xiang GU Yihang ZHAO Shuai XU Erhui WANG Dong GUO Yuejiang SHI Huasheng XIE Yunfeng LIANG Minsheng LIU the EHL-2 Team Hebei Key Laboratory of Compact Fusion Langfang 065001People’s Republic of China ENN Science and Technology Development Co. Ltd.Langfang 065001People’s Republic of China Centre for Theoretical and Computational Physics College of PhysicsQingdao UniversityQingdao 266071People’s Republic of China College of Computer Science South-Central Minzu UniversityWuhan 430074People’s Republic of China Key Laboratory of Materials Modification by Laser Ion and Electron Beams(Ministry of Education)School of PhysicsDalian University of TechnologyDalian 116024People’s Republic of China Department of Plasma Physics and Fusion Engineering University of Science and Technology of ChinaHefei 230026People’s Republic of China College of Physics and Electronic Engineering Hengyang Normal UniversityHenyang 421002People’s Republic of China Forschungszentrum Jülich GmbH Institute of Fusion Energy and Nuclear Waste Management-PlasmaphysikJülich 52425Germany

The EHL-2 spherical torus is designed to demonstrate proton-boron(p-11B)fusion within a compact spherical *** planned heating system includes a negative ion-based neutral beam injection(N-NBI),two positive ion-based NBI systems(P-NBI),electron cyclotron resonance heating(ECRH),ion cyclotron resonance heating(ICRH),and high harmonic fast wave(HHFW),with a total power output of 31 *** to scaling law estimates,the device is capable of achieving H-mode *** plasma density,,n_(e,min)at the minimum L-H power threshold,P_(lh),is estimated to be 4.4×10^(19)m^(-3).The pedestal parameters were calculated using the REPED *** B as the primary impurity ion,the predicted pedestal width and height are lower compared to the typical case with carbon *** pedestal collisionality for EHL-2 is estimated to range between 0.06 and 0.17,indicating the potential for significant energy loss due to edge localized modes(ELMs).The heat flux on the divertor plate has been calculated using the JOREK *** peak heat fluxes during ELM bursts are approximately 31.0 MW/m^(2)at the lower inboard target and 39.5 MW/m^(2)at the lower outboard target.A preliminary design of the resonant magnetic perturbation(RMP)coils has been completed to both control type-I ELMs and correct error *** system comprises 16 coils arranged into 24 *** ELM control mode,a 14/2 component is generated at 1.7 G/kAt,with a current of 4.9 kA required to achieveσChirikow=1 at the resonant surface,where the normalized poloidal magnetic flux is *** error field(EF)modulation mode,2/1 and 3/1 components are generated at 3.5 G/kAt and 2.8 G/kAt,respectively.

关键词： H-mode spherical torus(ST) EHL-2 pedestal resonant magnetic perturbations

来源：评论

学校读者我要写书评

暂无评论

Enhancing Energy Efficiency in Wireless Sensor Networks using Deep Learning

Enhancing Energy Efficiency in Wireless Sensor Networks usin...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Raskar, Sandeep Dhasmana, Gulshan Lakshminarayana, M. Patil, Harshal Bhagya Shree, S. Natrayan, L. Terna Engineering College Department of Artificial Intelligence and Data Science Nerul Maharashtra Navi Mumbai400706 India Graphic Era Deemed to be University Department of Computer Science and Engineering Uttarakhand Dehradun248002 India M S Ramaiah Institute of Technology Department of Medical Electronics Engineering Karnataka Bengaluru560054 India School of Computer Science & Engineering Iilm University Uttar Pradesh Greater Noida201 306 India Brindavan College of Engineering Department of Physics Karnataka Bengaluru560063 India Saveetha School of Engineering Simats Department of Mechanical Engineering Tamil Nadu Chennai602105 India

ISBN: (纸本)9798331509828

Wireless Sensor Networks (WSNs) are essential for many applications, but because of their limited power resources, they confront major energy efficiency difficulties. This research offers a novel approach for improving WSN energy efficiency that makes use of deep learning, namely deep reinforcement learning. The framework uses sophisticated energy data transformation methods to pre-process sensor data, derive important information, and spot energy usage trends. In order to better comprehend the model and represent the distinct dynamics of WSNs, domain-specific features including residual energy, transmission power, and path length are designed. The system learns the best routing and energy-saving strategies by incorporating deep reinforcement learning, which strikes a balance between energy usage and data transmission efficiency. When compared to traditional techniques, evaluation metrics show that the suggested framework can greatly reduce energy depletion and increase network longevity. Deep learning has the ability to address WSN energy concerns by providing a scalable and adaptable solution for energy-efficient network operations. © 2025 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Heralded optical entanglement distribution via lossy quantum channels: a comparative study

引用

Optics Express 2025年第6期33卷 12459-12474页

作者： Zo, Wan Chin, Seungbeom Kim, Yong-Su Seoul02792 Korea Republic of Department of Physics and Astronomy Seoul National University Seoul08826 Korea Republic of Department of Electrical and Computer Engineering Sungkyunkwan University Suwon16419 Korea Republic of Okinawa Institute of Science and Technology Graduate University Okinawa904-0495 Japan Division of Quantum Information Korea University of Science and Technology KIST School Seoul02792 Korea Republic of

Quantum entanglement serves as a foundational resource for various quantum technologies. In optical systems, entanglement distribution relies on the indistinguishability and spatial overlap of photons. Heralded schemes play a crucial role in ensuring the reliability of entanglement generation by detecting ancillary photons to signal the creation of desired entangled states. However, photon losses in quantum channels remain a significant challenge, limiting the distance and capacity of entanglement distributions. This study suggests three heralded schemes for distributing multipartite Greenberger-Horne-Zeilinger (GHZ) states via lossy quantum channels. These schemes differ both qualitatively in their network architecture (centralized or decentralized) and photon source requirements (Bell states or single-photons), and quantitatively in their success probabilities and heralding efficiency. Through comprehensive analysis incorporating both practical implementation considerations and theoretical performance metrics, we find that each scheme offers distinct advantages depending on the number of parties, channel distance, and security requirements. The decentralized scheme proves particularly advantageous for networks requiring balanced information distribution, while centralized schemes may offer better performance for smaller networks. This analysis provides insights into designing resilient heralded circuits for quantum information processing over lossy channels, considering both architectural constraints and performance requirements. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.

关键词： Quantum channel

来源：评论

学校读者我要写书评

暂无评论

Low Chirp Dual-Racetrack Modulator for PAM-4 Modulation with Low Driving Voltage

引用

Journal of Lightwave Technology 2025年第10期43卷 4840-4847页

作者： Yu, Fuhao Zeng, Zhaobang Tang, Kaifei Yang, Shichong Ji, Xiang Xin, Yu Gu, Tingyi Jiang, Wei Nanjing University Key Laboratory of Intelligent Optical Sensing and Manipulation Engineering Research Center for Precision Photonics Integration & System Applications Ministry of Education Nanjing210023 China Nanjing University National Laboratory of Solid-State Microstructures College of Engineering and Applied Sciences Institute of Optical Communication Engineering Nanjing210023 China Nanjing University of Information Science & Technology School of Physics and Optoelectronic Engineering Nanjing210044 China University of Delaware Department of Electrical and Computer Engineering NewarkDE19716 United States

Micro-ring modulators with a small footprint and energy consumption have been explored for densely integrated high throughput optical interconnects. However, their strong dispersive response results in frequency chirp and pulse broadening, which limits their signal transmission distance. Here we explore the low chirp operation of a dual-racetrack modulator symmetrically coupled to a bus waveguide. With opposite index changes in the two resonators, the coherent couplings of two resonators' optical fields with the bus waveguide and their interference can cancel out the output phase variation due to a single resonator. This allows the push-pull operation to achieve a lower chirp compared to the single-drive operation. The advantage of lower chirp operation is experimentally verified through open eye diagram with 100 Gbps PAM-4 transmitted over 2 km single-mode fiber (SMF), with a peak-to-peak driving voltage of 1.6 V. © 1983-2012 IEEE.

关键词： Single mode fibers

来源：评论

学校读者我要写书评

暂无评论

Bi2F-YOLO: a novel framework for underwater object detection based on YOLOv7

引用

Intelligent Marine Technology and Systems 2025年第1期3卷 1-12页

作者： Liu, Xiaopeng Zhao, Keke Liu, Cong Chen, Long College of Computer Science and Engineering Shandong University of Science and Technology Qingdao China School of Science Nanjing University of Science and Technology Nanjing China Department of Medical Physics and Biomedical Engineering University College London London UK

Underwater object detection faces significant challenges, including ambiguity and occlusion, which greatly undermine the accuracy of traditional algorithms. To address these issues, we propose Bi2F-YOLO, an algorithm, specifically designed for underwater environments. Bi2F-YOLO integrates the BiFormer module into the YOLOv7 backbone, utilizing Bi-Level routing attention (BRA) to focus on key features such as object edges and textures. This effectively addresses the problem of object ambiguity. In the detection head, we replace the conventional ELAN component with the FasterNet module. This update enhances detection efficiency and accuracy through the use of partial convolution (PConv), which redistributes the convolution kernel weights based on the sparsity of the input feature map. By doing so, it prevents the dilution of critical underwater object features caused by interference from irrelevant data. This effectively resolves the occlusion problem in underwater target detection while simultaneously reducing model parameters and computational costs. The experimental results show that Bi2F-YOLO achieves 87.3% $${{mAP}}_{50}$$ and 86.8% $${{AP}}_{50}$$ on the RUOD dataset. Compared to existing methods, it reduces model parameters by 14.3% fewer parameters (Params) and lowers computational costs by 5.5%, highlighting its efficiency and accuracy in underwater object detection.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Low-Complexity Channel Estimation for RIS-Assisted ISAC System

引用

China Communications 2025年第4期22卷 296-308页

作者： Chen Zhen Li Jianqing Zhang Haijun Zhang Wei The Institute of Microelectronics University of MacaoMacao 999078China School of Computer Science and Engineering Macao University of Science and TechnologyMacao 999078China Beijing Engineering and Technology Research Center for Convergence Networks and Ubiquitous Services University of Science and Technology BeijingBeijing 100083China Henan Academy of Science Applied Physics Institute Co. LtdZhengzhou 450008China

Integrated sensing and communication(ISAC),assisted by reconfigurable intelligent surface(RIS)has emerged as a breakthrough technology to improve the capacity and reliability of 6G wireless ***,a significant challenge in RIS-ISAC systems is the acquisition of channel state information(CSI),largely due to co-channel interference,which hinders meeting the required reliability *** address this issue,a minimax-concave penalty(MCP)-based CSI refinement scheme is *** approach utilizes an element-grouping strategy to jointly estimate the ISAC channel and the RIS phase shift *** previous methods,our scheme exploits the inherent sparsity in RIS-assisted ISAC channels to reduce training overhead,and the near-optimal solution is derived for our studied RIS-ISAC *** effectiveness of the element-grouping strategy is validated through simulation experiments,demonstrating superior channel estimation results when compared to existing benchmarks.

关键词： channel estimation element grouping integrated sensing and communication(ISAC) RIS

来源：评论

学校读者我要写书评

暂无评论

Predictive modeling of MXene-based solar absorbers using a deep neural network

引用

Journal of the Optical Society of America B: Optical physics 2025年第4期42卷 763-772页

作者： Waseer, Waleed Iqbal Baqir, Muhammad Abuzar Saqlain, Muhammad Mughal, Muhammad Junaid Khan, Shujaat School of Physics State Key Laboratory of Electronic Thin Films and Integrated Devices University of Electronic Science and Technology of China Chengdu China Department of Electrical Engineering COMSATS University Islamabad Sahiwal57000 Pakistan Department of Electrical and Computer Engineering COMSATS University Islamabad Islamabad45550 Pakistan Department of Computer Engineering College of Computing and Mathematics King Fahd University of Petroleum & Minerals Dhahran31261 Saudi Arabia SDAIA-KFUPM Joint Research Center for Artificial Intelligence King Fahd University of Petroleum & Minerals Dhahran31261 Saudi Arabia

Nanophotonic structures, such as photonic crystals, plasmonic nanostructures, and metamaterials, present transformative potential in advancing optical devices through innovative design capabilities. Among these, metamaterials—specifically metal–insulator–metal (MIM)-based absorbers—stand out for their ability to achieve high electromagnetic wave absorption across designated frequencies, proving valuable in applications ranging from solar cells to sensors. This study focuses on designing a high-efficiency MXene–SiO2–silver MIM metasurface absorber, leveraging deep learning techniques to streamline and enhance the simulation process traditionally reliant on finite-difference time-domain (FDTD) or finite-element methods (FEMs). By utilizing a deformable convolutional neural network (CNN), the proposed model predicts absorption spectra from metasurface geometries with superior accuracy and reduced computational demand while achieving dynamic spatial transformation handling. The model’s accuracy is further elucidated through Gradient-weighted Class Activation Mapping (Grad-CAM), which provides insights into the absorber’s most influential regions, supporting targeted optimization for specific applications. This deep learning approach demonstrates significant advantages over conventional methods in efficiency and adaptability, advancing the development of high-performance nanophotonic devices. © 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.

关键词： Time domain analysis

来源：评论

学校读者我要写书评

暂无评论

A technical assessment of latent diffusion for Alzheimer's Disease progression

A technical assessment of latent diffusion for Alzheimer's D...

引用

Medical Imaging 2025: Image Processing

作者： McMaster, Elyssa Puglisi, Lemuel Gao, Chenyu Krishnan, Aravind R. Saunders, Adam M. Ravi, Daniele Beason-Held, Lori L. Resnick, Susan M. Zuo, Lianrui Moyer, Daniel Landman, Bennett A. Department of Electrical and Computer Engineering Vanderbilt University NashvilleTN United States Department of Computer Science University of Catania Catania Italy School of Physics Engineering and CS University of Hertfordshire United Kingdom Laboratory of Behavioral Neuroscience National Institute on Aging National Institutes of Health BaltimoreMD United States Department of Computer Science Vanderbilt University NashvilleTN United States

ISBN: (纸本)9781510685901

Latent Diffusion Models (LDMs) introduce exciting opportunities in medical imaging, from disease progression prediction to interpolation to generate entire datasets of rare data. The stochastic nature of generative models makes it challenging to validate their outputs and assess their robustness across diverse datasets. BrLP, a state-of-the-art LDM for T1-weighted (T1w) images that incorporates auxiliary brain volume information, has been evaluated on Alzheimer's Disease (AD) progression, and achieves structural similarity index (SSIM) of 0.91 ± 0.03. In this work, we conducted a pilot study of the BrLP model using the Baltimore Longitudinal Study of Aging (BLSA) dataset. Our objectives are to (1) evaluate the model performance on an external dataset using pretrained image-based and brain image-based metrics such as mean squared error (MSE), similarity index, and mean absolute error (MAE) between conditional and unconditional brain regions;and (2) determine if a harmonization step, in addition to the proposed model's preprocessing steps, is required to improve performance. We found that the BrLP is robust to T1w imaging scanner effects, and harmonization is not required. However, there exists a bias toward a younger population compared to the BLSA cohort (BrLP reported 0.91 ± 0.03 SSIM in a cohort with age range 74 ±7 years;SSIM in our experiment is 0.91 ± 0.03 in cognitively normal subjects with age range 79.47 years ± 7.35 and AD cohort 0.90 ± 0.012 with age range 83.73 years ± 6.01). Interestingly, when the model's input was changed to simulate AD progression instead of normal aging, a higher SSIM of 0.91 ± 0.0012 was achieved compared to a ground truth non-AD scan, suggesting a potential mismatch. However, the resulting lower conditional volume regions function as expected. The model's architecture shows promise for longitudinal T1w imaging studies. © 2025 SPIE.

关键词： Neurodegenerative diseases

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：