检索结果-内蒙古大学图书馆

Numerical-experimental solar loading thermography method and chemical-physical techniques: A Tandem for the preservation of ancient books

引用

Journal of Cultural Heritage 2025年 73卷 579-589页

作者： Gilda Russo Yinuo Ding Reagan Kasonsa Tshiangomba Enza Pellegrino Antonio Cicone Hai Zhang Stefano Sfarra Department of Industrial and Information Engineering and Economics University of L'Aquila L’Aquila I-67100 Italy Centre for Composite Materials and Structures (CCMS) Harbin Institute of Technology Harbin 150001 China Department of Information Engineering Computer Science and Mathematics University of L'Aquila L’Aquila I-67100 Italy

In recent decades, the need to preliminarily study works of art with non-destructive and portable techniques has given rise to the figure of the conservation scientist for applications in the field of cultural heritage. This study applies solar loading thermography to detect surface and subsurface defects in an ancient book (1861), examining both natural degradation and fabricated defects. The latter were generated by the natural and inevitable degradation process to which the organic components of the book (for example cellulose, lignin, etc.) are subjected, and voluntarily introduced inside the book cover to determine the sensitivity and the feasibility of the technique. Thermal imaging analysis, supported by numerical simulation, revealed humidity damage and adhesive residues. Two experimental conditions were tested using or not clips to optimize cover-to-page adhesion. Four circoular dowels of different compositions assessed technique sensitivity. Complementary analyses (UV fluorescence, XRF spectroscopy, optical microscopy) validated surface anomal detection and material characterization.

关键词： Infrared thermography Ancient book Optical microscope Ultraviolet fluorescence XRF spectroscopy Numerical simulation

来源：评论

学校读者我要写书评

暂无评论

SwimVG: Step-wise Multimodal Fusion and Adaption for Visual Grounding

arXiv

引用

arXiv 2025年

作者： Shi, Liangtao Liu, Ting Hu, Xiantao Hu, Yue Yin, Quanjun Hong, Richang Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Hefei230009 China Ministry of Education and School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China School of systems engineering National University of Defense Technology Hunan Province Changsha410073 China Department of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210014 China

Visual grounding aims to ground an image region through natural language, which heavily relies on cross-modal alignment. Most existing methods transfer visual/linguistic knowledge separately by fully fine-tuning uni-modal pre-trained models, followed by a simple stack of visual-language transformers for multimodal fusion. However, these approaches not only limit adequate interaction between visual and linguistic contexts, but also incur significant computational costs. Therefore, to address these issues, we explore a step-wise multimodal fusion and adaption framework, namely SwimVG. Specifically, SwimVG proposes step-wise multimodal prompts (Swip) and cross-modal interactive adapters (CIA) for visual grounding, replacing the cumbersome transformer stacks for multimodal fusion. Swip can improve the alignment between the vision and language representations step by step, in a token-level fusion manner. In addition, weight-level CIA further promotes multimodal fusion by cross-modal interaction. Swip and CIA are both parameter-efficient paradigms, and they fuse the cross-modal features from shallow to deep layers gradually. Experimental results on four widely-used benchmarks demonstrate that SwimVG achieves remarkable abilities and considerable benefits in terms of efficiency. Our code is available at https://***/liuting20/SwimVG. Copyright © 2025, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

A Novel Automatic Text Document Classification Using Learning based Text Classification(LbTC) Approach

引用

Procedia computer science 2025年 258卷 4279-4290页

作者： Avinash N J Krishnaraj Rao Rama Moorthy H Raviprakash B Raghunadan K R Vasudeva Venkatadri M Department of Electronics and Communication Engineering Mangalore Institute of Technology and Engineering Moodabidre 574225 Karnataka India Department of Information Science and Engineering NMAM Institute of Technology Nitte (Deemed to be University) Karkala 574110 Karnataka India Department of Computer Applications Nitte Institute of Professional Education Nitte (Deemed to be University) Mangalore 575007 Karnataka India Department of Computer Science and Engineering NMAM Institute of Technology Nitte (Deemed to be University) Karkala 574110 Karnataka India Associate Dean MPSTME Shirpur Campus NMIMS (Deemed-to-be-University) Shirpur Maharashtra India

There are practical uses for text document classification. In actuality, it is crucial to categorize and parse documents with natural language. Utilizing applications like bogus news identification, query tagging, sentiment classification, and spam filtering needs this sort of study. However, because of their ambiguity, open-ended nature, and vastness, text documents provide a difficult Classification challenge. Machine learning (ML) methods became valuable alongside the growth of artificial intelligence (AI) due to their data-driven learning techniques. These methods are seen as highly effective at handling and analyzing vast data sets in depth. Topic segmentation, text categorization, entity identification, machine translation, and text summarization, to name a few, are just a few of the issues that may be resolved with ML approaches. In this study, we introduced an automatic Text Classification system (ATCF), a system that uses shallow and deep neural networks to classify text documents. To implement our system, we proposed an approach called Learning based Text Classification (LbTC). We investigated the performance of several models in comparison. Based on the training provided to ML models, the suggested framework assists in categorizing any type of document. It is compatible with practical applications where the categorization of documents is essential.

关键词： Machine learning shallow learning deep learning text classification document Classification

来源：评论

学校读者我要写书评

暂无评论

A multi-task neural network for full waveform ultrasonic bone imaging

引用

computer Methods and Programs in Biomedicine 2025年 267卷 108807-108807页

作者： Li, Peiwen Liu, Tianyu Ma, Heyu Li, Dan Liu, Chengcheng Ta, Dean Department of Biomedical Engineering School of Information Science and Technology Fudan University Shanghai200438 China Institute of Biomedical Engineering & Technology Academy for Engineering and Technology Fudan University Shanghai200438 China Department of Electronic Engineering School of Information Science and Technology Fudan University Shanghai200438 China State Key Laboratory of Integrated Chips and Systems Fudan University Shanghai201203 China

Background and objective: It is a challenging task to use ultrasound for bone imaging, as the bone tissue has a complex structure with high acoustic impedance and speed-of-sound (SOS). Recently, full waveform inversion (FWI) has shown promising imaging for musculoskeletal tissues. However, the FWI showed a limited ability and tended to produce artifacts in bone imaging because the inversion process would be more easily trapped in local minimum for bone tissue with a large discrepancy in SOS distribution between bony and soft tissues. In addition, the application of FWI required a high computational burden and relatively long iterations. The objective of this study was to achieve high-resolution ultrasonic imaging of bone using a deep learning-based FWI approach. Method: In this paper, we proposed a novel network named CEDD-Unet. The CEDD-Unet adopts a Dual-Decoder architecture, with the first decoder tasked with reconstructing the SOS model, and the second decoder tasked with finding the main boundaries between bony and soft tissues. To effectively capture multi-scale spatial-temporal features from ultrasound radio frequency (RF) signals, we integrated a Convolutional LSTM (ConvLSTM) module. Additionally, an Efficient Multi-scale Attention (EMA) module was incorporated into the encoder to enhance feature representation and improve reconstruction accuracy. Results: Using the ultrasonic imaging modality with a ring array transducer, the performance of CEDD-Unet was tested on the SOS model datasets from human bones (noted as Dataset1) and mouse bones (noted as Dataset2), and compared with three classic reconstruction architectures (Unet, Unet++, and Att-Unet), four state-of-the-art architecture (InversionNet, DD-Net, UPFWI, and DEFE-Unet). Experiments showed that CEDD-Unet outperforms all competing methods, achieving the lowest MAE of 23.30 on Dataset1 and 25.29 on Dataset2, the highest SSIM of 0.9702 on Dataset1 and 0.9550 on Dataset2, and the highest PSNR of 30.60 dB

关键词： Ultrasonography

来源：评论

学校读者我要写书评

暂无评论

Machine learning and Big Data in deep underground engineering

引用

Deep Underground science and engineering 2025年第1期4卷 1-2页

作者： Asoke K.Nandi Ru Zhang Tao Zhao Tao Lei Department of Electronic and Electrical Engineering Brunel University of LondonUxbridgeUK College of Water Resource and Hydropower Sichuan UniversityChengduChina Institute of Rock and Soil Mechanics Chinese Academy of SciencesWuhanChina School of Electronic Information and Artificial Intelligence Shaanxi University of Science and TechnologyXi'anChina

This special issue of Deep Underground science and engineering(DUSE)showcases pioneering research on the transformative role of machine learning(ML)and Big Data in deep underground *** by guest editors *** Nandi(Brunel University of London,UK),*** Zhang(Sichuan University,China),*** Zhao(Chinese Academy of sciences,China),and *** Lei(Shaanxi University of science and technology,China),this issue highlights the innovative applications of ML technique in reshaping structural safety,tunneling operations,and geotechnical investigations.

关键词： underground learning operations

来源：评论

学校读者我要写书评

暂无评论

Asynchronism in Cellular Automata

arXiv

引用

arXiv 2025年

作者： Gautam, Virendra Kumar Department of Information Technology Indian Institute of Engineering Science and Technology Shibpur West Bengal Howrah711103 India

This study presents the concept of Skewed Fully Asynchronous Update in cellular automata and examines the resulting behaviour. It also investigates the dynamics of elementary cellular automata under this update scheme, comparing it with other update methods such as synchronous and fully asynchronous updates. Furthermore, the research introduces the application of fully asynchronous cellular automata for solving clustering problems and explores the dynamics of elementary cellular automata within the framework of α-asynchronous cellular automata. In this work, we introduce a new type of asynchronous cellular automata called Skewed Fully Asynchronous Cellular Automata (SACA) as a source of noise in cellular auotmata. The Skewed Fully asynchronous cellular automata allows to update the states of only two consecutive and adjacent cells, say ci and ci+1, simultaneously at each and every time step. Under this proposed scheme, we analyzed dynamics and behaviour of elementary cellular automata (ECA). The dynamical behaviour are compared with the fully asynchronous cellular automata, and synchronous cellular automata. This comparative study points out varieties of rich phenomenon that some of the elementary cellular automata shift from convergent nature (or non reversible non convergent dynamics) to reversible nature (or divergence). We also identify the cases where the divisibility of the lattice size by 2 or 4 introduces massive repercussion in the system following presence or absence of the atomicity property. Lastly, we theorize the reason behind convergence towards all 0 and all 1 point attractors under the proposed skewed environment which partially validates our experimental observations. The work also presents the study and outcomes of the application of fully asynchronous cellular automata for solving clustering problem. By clustering, we mean a group of objects having similar characteristics (or properties). This work uses reversible asynchronous cellular automata.

关键词： Automata theory

来源：评论

学校读者我要写书评

暂无评论

Near-Field Communications: Shape and Structure Design for Uniform Planar Array

引用

IEEE Internet of Things Journal 2025年

作者： Chen, Siyu Gu, Juping Duan, Wei Zhang, Lei Dang, Shuping Wen, Miaowen Ding, Zhiguo Ho, Pin-Han Nantong University School of Information Science and Technology Nantong226019 China Suzhou University of Science and Technology China Hubei Polytechnic University Department of Electronics and Information Engineering Huangshi435003 China University of Bristol School of Electrical Electronic and Mechanical Engineering BristolBS8 1UB United Kingdom South China University of Technology School of Electronic and Information Engineering Guangzhou510640 China Khalifa University Department of Electrical Engineering and Computer Science Abu Dhabi United Arab Emirates University of Electronic Science and Technology of China Shenzhen Institute for Advanced Study Shenzhen518000 China University of Waterloo Department of Electrical and Computer Engineering WaterlooONN2L 3G1 Canada

With the flourishing development of 6G wireless networks, the demand of spectrum efficiency rapidly increases, in order to support the demands of high quality data transmission and connections of massive users. Among various promising technologies, the technology of near-field communications provides a great potential to address such an issue due to the unique spherical-wave channels for electromagnetic (EM) propagation. In this paper, multiple-input single-output (MISO) near-field communications is comprehensively studied to clarify the influences of the shape and structure of uniform planar array (UPA) on the system performance, considering three cases with uniform linear array (ULA), rectangular UPA, and circular UPA. In particular, we reveal the properties of rapid deterioration for signal-to-noise ratio (SNR) from the reduced projection aperture in near-field communications and investigate the single spherical crown antenna design and spherical crown antenna array design in order to address this issue. Moreover, we also characterize the role of antenna projection aperture in details, and theoretically analyze the shape of UPA, yielding the corresponding exact closed-form expressions for SNR and outage probability (OP). Based on these above analytical results, we find out that adjusting the spacing between adjacent antennas to control the relative angle between user and selected antennas is an efficient way to improve the projection aperture of antenna and SNR. Simulation results are shown to well match analytical results, which validate the correctness of our analysis, clarifying that our proposed designs outperform the conventional works and illustrating a better stability for angle variations. © 2025 IEEE.

关键词： Spheres

来源：评论

学校读者我要写书评

暂无评论

Blockchain for Machine Learning: Security and Trust in Data Sharing

Blockchain for Machine Learning: Security and Trust in Data ...

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Kiran, G Ravi Maisaiah, Nakka Divya, T.V. Reddy, G V Subba Petikam, Sailaja Sharma, M.K. Sr University School of Cs & Ai Telangana Warangal India Mahaveer Institute of Science and Technology Department of Information Technology Telangana Hyderabad500005 India Sreyas Institute of Engineering and Technology Department of Cse Telangana Hyderabad500068 India Siddhartha Institute of Technology and Science Ece Department Hyderabad India Simats Saveetha School of Law Chennai600077 India Mathematics Chaudhary Charan Singh University Uttar Pradesh Meerut250004 India

ISBN: (纸本)9798331508685

Data security has become the most important things in this current digitalized environment. As in this era, each process leaves a digital footprint and it requires significant monitoring to maintain trust of stakeholders. It necessitates the companies to evaluate their necessary information including the sensitive and confidential data as per their requirements while maintaining data security and customers' trustworthiness. While managing data security, companies require limited accessibility of their confidential business information. Personal and private information are the most commonly targeted dataset for hackers in this technological age. Since ML is efficient in evaluating huge datasets, companies must prioritize the potential security concerns. To use artificial intelligence (AI) and ML-driven technologies widely, blockchain has become significant to develop faith among the stakeholders. The research evaluates the efficiency of blockchain technology in improving data security while suggesting its integration with ML-based business approaches. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Threat Detection: Implementing Multi-Layer Security Networks in Cloud Environments

AI-Driven Threat Detection: Implementing Multi-Layer Securit...

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Surya, S. Santhakumar, D. Tyagi, Anuj Onapakala, Kiran Thurai Raaj, V.B. Krishna Kumar, N. Chennai600116 India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Computer Science and Engineering Tamil Nadu Chennai602105 India RingCentral Inc Sr. Site Reliability Eng United States PacificSource Health Plans PortlandOR United States Madanapalle Institute of Technology & Science Electrical and Electronics Engineering Andhra Pradesh Madanapalle India Guru Nanak University University Institute of Computer Science and Applications Department of Csa Telangana Hyderabad501506 India

ISBN: (纸本)9798331508685

AI-driven threat detection has become a vital tool for security maintenance as cloud environments become more and more integrated into digital infrastructures. In order to identify and reduce risks in cloud-based systems, this study investigates the deployment of multi-layer security networks utilising cutting-edge machine learning (ML) and deep learning (DL) models. We created a hybrid training strategy that blends historical and synthetic data for improved relevance in contemporary cloud environments by utilising datasets like KDD Cup 99 and CSECIC-IDS2018 in addition to simulated cloud network interactions. ML models such as Random Forest and Logistic Regression were used to create baseline threat detection, while Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks were used to accomplish deeper pattern recognition. The effectiveness of DL models in proactively protecting cloud infrastructures against changing cyber threats was demonstrated by thorough performance evaluations based on accuracy, precision, recall, F1 Score, and AUC-ROC. These evaluations also demonstrated the models' better flexibility in dynamic cloud environments. © 2025 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Bibliometric and Content Analysis of Large Language Models Research in Software engineering: The Potential and Limitation in Software engineering

引用

International Journal of Advanced computer science and Applications 2025年第4期16卷 344-356页

作者： Damayanti, Annisa Dwi Gani, Hamdan Zhipeng, Feng Gani, Helmy Zuhriyah, Sitti Nurani Djabir, St. Nurhayati Wardani, Nur Ilmiyanti Department of Environmental Engineering Faculty of Engineering Hasanuddin University Makassar Indonesia Department of Machinery Automation System ATI Makassar Polytechnic Makassar Indonesia School of Culture Creativity and Media Hangzhou Normal University Zhejiang Hangzhou China Department of Industrial Hygiene Faculty of Public Health Occupational Health and Safety Makassar College of Health Sciences Indonesia Department of Computer System Universitas Handayani Makassar Makassar Indonesia Department of Information Systems and Technology Institut Teknologi dan Bisnis Nobel Indonesia Jl. Sultan Alauddin No.212 Makassar90221 Indonesia Department of Informatics Engineering Universitas Handayani Makassar Makassar Indonesia

Large Language Models (LLM) is a type of artificial neural network that excels at language-related tasks. The advantages and disadvantages of using LLM in software engineering are still being debated, but it is a tool that can be utilized in software engineering. This study aimed to analyze LLM studies in software engineering using bibliometric and content analysis. The study data were retrieved from Web of science and Scopus. The data were analyzed using two popular bibliometric approaches: bibliometric and content analysis. VOS Viewer and Bibliometrix software were used to conduct the bibliometric analysis. The bibliometric analysis was performed using science mapping and performance analysis approaches. Various bibliometric data, including the most frequently referenced publications, journals, and nations, were evaluated and presented. Then, the synthetic knowledge method was utilized for content analysis. This study examined 235 papers, with 836 authors contributing. The publications were published in 123 different journals. The average number of citations per publication is 1.44. Most publications were published in Proceedings International Conference on Software engineering and ACM International Conference Proceeding Series, with China and the United States emerging as the leading countries. It was discovered that international collaboration on the issue was inadequate. The most often used keywords in the publications were "software design," "code (symbols)," and "code generation." Following the content analysis, three themes emerged: 1) Integration of LLM into software engineering education, 2) application of LLM in software engineering, and 3) potential and limitation of LLM in software engineering. The results of this study are expected to provide researchers and academics with insights into the current state of LLM in software engineering research, allowing them to develop future conclusions. © (2025), (science and information Organization). All Rights Res

关键词： Software design

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：