检索结果-内蒙古大学图书馆

Cross-Modal Recipe Retrieval With Fine-Grained Prompting Alignment and Evidential Semantic Consistency

IEEE Transactions on Multimedia 2025年 27卷 2783-2794页

作者： Huang, Xu Liu, Jin Zhang, Zhizhong Xie, Yuan Tang, Yongqiang Zhang, Wensheng Cui, Xiaohui Wuhan University School of Computer Science Wuhan430072 China Ministry of Education Wuhan430062 China East China Normal University School of Computer Science and Technology Shanghai200062 China Institute of Automation Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Beijing100190 China Wuhan University School of Cyber Science and Engineering Wuhan430072 China

Alignment between the food images and the corresponding recipes is an emerging cross-modal representation learning task. In this task, the recipes are composed of three components, i.e., food title, ingredient lists, and cooking instructions, which require a fine-grained alignment between the features of the two modalities. Existing methods usually aggregate the recipes into global embeddings and then align them with the global image embeddings. Meanwhile, semantic classification is frequently used in these methods to regularize the embeddings of the two modalities. While these methods are efficient, there remain two problems. (1) Forcing the alignment between the global images and recipes embeddings may result in losing the component-specific information. (2) The high diversity of food appearance leads to high uncertainty in the semantic classification of food images and recipes. To solve these problems, we propose a Fine-grained Prompting and Alignment (FPA) model to enhance the feature extraction and bring more component-specific information for fine-grained alignment. Furthermore, to regularize the semantic information contained in the cross-modal features, we design an Evidential Semantic Consistency (ESC) loss to keep the cross-modal semantic consistency. We have conducted comprehensive experiments on the benchmark dataset Recipe1M and the state-of-the-art results on the cross-modal recipe retrieval task demonstrate the effectiveness of our method. © 1999-2012 IEEE.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Breast Mammogram Analysis and Classification Using Deep Convolution Neural Network

引用

computer systems Science & engineering 2022年第10期43卷 275-289页

作者： V.Ulagamuthalvi G.Kulanthaivel A.Balasundaram Arun Kumar Sivaraman Department of Computer Science and Engineering Sathyabama UniversityChennai600119India Department of Electrical Electronics and Communication EngineeringNITTTRChennai600113India Centre for Cyber Physical Systems School of Computer Science and EngineeringVellore Institute of Technology(VIT)Chennai600127India School of Computer Science and Engineering Vellore Institute of Technology(VIT)Chennai600127India

One of the fast-growing disease affecting women’s health seriously is breast *** is highly essential to identify and detect breast cancer in the earlier *** paper used a novel advanced methodology than machine learning algorithms such as Deep learning algorithms to classify breast cancer *** learning algorithms are fully automatic in learning,extracting,and classifying the features and are highly suitable for any image,from natural to medical *** methods focused on using various conventional and machine learning methods for processing natural and medical *** is inadequate for the image where the coarse structure matters *** of the input images are downscaled,where it is impossible to fetch all the hidden details to reach accuracy in *** deep learning algorithms are high efficiency,fully automatic,have more learning capability using more hidden layers,fetch as much as possible hidden information from the input images,and provide an accurate *** this paper uses AlexNet from a deep convolution neural network for classifying breast cancer in mammogram *** performance of the proposed convolution network structure is evaluated by comparing it with the existing algorithms.

关键词： Medical image processing deep learning convolution neural network breast cancer feature extraction classification

来源：评论

学校读者我要写书评

暂无评论

A Reasoning Engine Architecture for Building Energy Metadata Management 13

A Reasoning Engine Architecture for Building Energy Metadata...

引用

13th International Conference on Information, Intelligence, systems and Applications, IISA 2022

作者： Kapsalis, Panagiotis Kormpakis, Giorgos Alexakis, Konstantinos Karakolis, Evangelos Mouzakitis, Spiros Askounis, Dimitris National Technical University of Athens Decision Support Systems Laboratory School of Electrical and Computer Engineering Athens Greece

ISBN: (数字)9781665463904

ISBN: (纸本)9781665463904

During the Buildings' lifecycle, massive amounts of data, that contain information related to their energy consumption, are generated. Towards the creation of smart building networks, this produced information must be intercepted and harmonized according to building ontologies and schemas. The pattern recognition from building metadata is based on inferencing and intelligent querying, that can be achieved with the utilization of graph and property databases that deploy and host building information. This paper presents a Reasoning Engine Architecture implemented in the context of the H2020 project called MATRYCS that persists building semantic information. It will be leveraged to support real life applications by improving the inference operations. © 2022 IEEE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

A hybrid blockchain and machine learning approach for intrusion detection system in Industrial Internet of Things

引用

Alexandria engineering Journal 2025年 127卷 619-627页

作者： Song, Wu Zhu, Xiangyuan Ren, Sheng Tan, Wenxue Peng, Yibo School of Computer and Electrical Engineering Hunan University of Arts and Science Hunan Changde415000 China Hunan Provincial Key Laboratory of Network Investigational Technology Hunan Police Academy Hunan Changsha410138 China

The Industrial Internet of Things (IIoT) is a key component of Industry 4.0, which enables manufacturing to be automated and data collected in real-time. Edge IoT devices are subject to cybersecurity threats and unauthorised access. Decentralisation and resource limitations of IIoT often prevent traditional security mechanisms from addressing these threats. Intrusion detection systems (IDSs), which are used to detect intrusions in IIoT environments, are presented in this paper as hybrid machine learning-blockchain approaches. Blockchain technology ensures data integrity, secures communication, and prevents unauthorised modifications through the proposed system. To reduce false positives and improve threat detection accuracy, XGBoost is able to reduce the number of false positives. Using the BOT-IoT dataset, the model is demonstrated to be superior to conventional intrusion detection systems. This approach ensures enhanced security and trustworthiness of IIoT networks by offering a scalable, efficient, and secure solution. © 2025

关键词： Industrial internet of things (IIoT)

来源：评论

学校读者我要写书评

暂无评论

ENERGY TRANSITIONS, INTELLIGENCE AND BIG DATA: Towards a prosumer concept with energy autonomy 13

ENERGY TRANSITIONS, INTELLIGENCE AND BIG DATA: Towards a pro...

引用

13th International Conference on Information, Intelligence, systems and Applications, IISA 2022

作者： Doukas, Haris National Technical University of Athens Decision Support Systems Laboratory School of Electrical and Computer Engineering Athens Greece

ISBN: (数字)9781665463904

ISBN: (纸本)9781665463904

Energy transition refers to the transition of the global energy sector towards renewable energy sources and energy efficiency, replacing fossil-based energy production and consumption systems (oil, gas and coal). The four main pillars towards this energy transition, also known as 4 Ds, will be presented as follows: Decarbonization, Decentralization, Digitization and Democratization. These dimensions are combined to change the nature of the way energy is produced, purchased and sold. In this context, consumer-centric big data application services and applications are being developed and will be discussed, based on the large amount of information available and the combination of the evolution of artificial intelligence and 'big' data management technologies. Emphasis will be given to the optimal processing and use of data with the help of innovative techniques towards intelligent energy services. Such services can enable consumers to adopt energy-efficient practices, their transition to prosumers and the promotion of energy autonomy concepts, through the creation of resilient and energy-efficient communities, bringing multiple benefits to households and industry. © 2022 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Secure Communication With UAV-Enabled Aerial RIS: Learning Trajectory With Reflection Optimization

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2023年 1-10页

作者： Tang, Xiao Jiang, Tianqi Liu, Jinxin Li, Bin Zhai, Daosen Yu, F. Richard Han, Zhu School of Electronics and Information Northwestern Polytechinical University Xi'an China China Helicopter Research and Development Institute Jingdezhen China School of Mechanical Engineering Xi'an Jiaotong University Xi'an China School of Electronics and Information Northwestern Polytechnical University Xi'an China Department of Systems and Computer Engineering Carleton University Ottawa ON Canada Department of Electrical and Computer Engineering University of Houston Houston USA

Reconfigurable intelligent surfaces (RISs) have manifested huge potential in enhancing information security by actively intervening the wireless propagation, yet the security gain may still be limited depending on the RIS deployment. In this paper, we propose to employ an unmanned aerial vehicle (UAV) mounting a RIS to enable on-demand reflection, noted as an aerial RIS (ARIS). The ARIS is then exploited to assist the anti-eavesdropping communications established through a conventional fixed-deployed RIS to further enhance the wireless secrecy. The secure communication is investigated by jointly optimizing the reflection at both RISs as well as the trajectory of the ARIS to maximize the average secrecy rate during the flight. To facilitate effective algorithm design, the formulated security problem is decomposed and solved in a double-layer framework. The outer layer tackles the flying trajectory through deep reinforcement learning while the inner layer solves for reflection phase shift design with manifold optimization. Finally, simulation results demonstrate the learned trajectory in various topologies as well as the superior performance of our proposal in terms of security provisioning. IEEE

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Quantifying Mobile Ions in Formamidinium Lead Iodide Perovskite to Study Ion Migration for Enhanced Stability and Performance 52

Quantifying Mobile Ions in Formamidinium Lead Iodide Perovsk...

引用

52nd IEEE Photovoltaic Specialist Conference, PVSC 2024

作者： Purushothama, Hithesh Rai Penukula, Saivineeth Rolston, Nicholas School of Electrical Computer and Energy Engineering Arizona State University Renewable Energy Materials and Devices Laboratory TempeAZ85281 United States

ISBN: (纸本)9781665464260

In our work, we explore the efficacy of ionic additives in enhancing the stability and processability of formamidinium lead iodide perovskite solar cells. This class of materials, known for its performance and stability, is often hindered by phase transitions to non-semiconductor phases due to the large tolerance factor which presents unique processing difficulties. Leveraging small alkali cation additives along with a methylammonium chloride additive, we aim to facilitate crystallization and achieve a stable crystal structure while simultaneously mitigating ion migration. Our research addresses the significant issue of mobile ion concentration, a key obstacle in perovskite technology. The addition of alkali cation additives has proven instrumental in curtailing the concentration of mobile ions, a primary source of degradation and hysteresis in perovskite solar cells. This breakthrough not only enhances the overall stability of these cells but also contributes to improved morphology and optoelectronic properties based on photoluminescence responses from the films. Our study delves into the underlying mechanisms responsible for these enhancements, providing valuable insights and paving the way for the development of reliable, efficient, and manufacturable stable absorber chemistry toward perovskite commercialization. © 2024 IEEE.

关键词： Perovskite

来源：评论

学校读者我要写书评

暂无评论

Towards explainable evaluation of language models on the semantic similarity of visual concepts 29

Towards explainable evaluation of language models on the sem...

引用

29th International Conference on Computational Linguistics, COLING 2022

作者： Lymperaiou, Maria Manoliadis, George Mastromichalakis, Orfeas Menis Dervakos, Edmund G. Stamou, Giorgos Artificial Intelligence and Learning Systems Laboratory School of Electrical and Computer Engineering National Technical University of Athens Greece

Recent breakthroughs in NLP research, such as the advent of Transformer models have indisputably contributed to major advancements in several tasks. However, few works research robustness and explainability issues of their evaluation strategies. In this work, we examine the behavior of high-performing pre-trained language models, focusing on the task of semantic similarity for visual vocabularies. First, we address the need for explainable evaluation metrics, necessary for understanding the conceptual quality of retrieved instances. Our proposed metrics provide valuable insights in local and global level, showcasing the inabilities of widely used approaches. Secondly, adversarial interventions on salient query semantics expose vulnerabilities of opaque metrics and highlight patterns in learned linguistic representations. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Cloud-edge-based We-Market:Autonomous Bidding and Peer-to-peer Energy Sharing Among Prosumers

引用

Journal of Modern Power systems and Clean Energy 2023年第4期11卷 1282-1293页

作者： Fangyuan Si Yinghua Han Qinqin Xu Jinkuan Wang Qiang Zhao College of Information Science and Engineering Northeastern UniversityShenyangChinaand F.Si is also with the State Key Laboratory of Power System and Generation EquipmentDepartment of Electrical EngineeringTsinghua UniversityBeijingChina School of Computer and Communication Engineering Northeastern University at QinhuangdaoQinhuangdaoChina School of Control Engineering Northeastern University at QinhuangdaoQinhuangdaoChina

With the extensive penetration of distributed renewable energy and self-interested prosumers,the emerging power market tends to enable user autonomy by bottom-up control and distributed *** paper is devoted to solving the specific problems of distributed energy management and autonomous bidding and peer-to-peer(P2P)energy sharing among prosumers.A novel cloud-edge-based We-Market is presented,where the prosumers,as edge nodes with independent control,balance the electricity cost and thermal comfort by formulating a dynamic household energy management system(HEMS).Meanwhile,the autonomous bidding is initiated by prosumers via the modified Stone-Geary utility *** the cloud center,a distributed convergence bidding(CB)algorithm based on consistency criterion is developed,which promotes faster and fairer bidding through the interactive iteration with the edge ***,the proposed scheme is built on top of the commercial cloud platform with sufficiently secure and scalable computing *** results show the effectiveness and practicability of the proposed We-Market,which achieves 15%cost reduction with shorter running *** analysis indicates better scalability,which is more suitable for largerscale We-Market implementation.

关键词： We-Market bidding energy sharing prosumer peer-to-peer cloud-edge

来源：评论

学校读者我要写书评

暂无评论

4D-Rossler Hyperchaotic System for Image Encryption and Decryption with High Security 10

4D-Rossler Hyperchaotic System for Image Encryption and Decr...

引用

10th International Conference on Communication and Signal Processing, ICCSP 2024

作者： Agarwal, Soumil Byramdharmika, M. Dhathri, Sree Thanikaiselvan, V. Subashanthini, S. Amirtharajan, Rengarajan Vellore Institute of Technology School of Electronics Engineering Vellore India Engineering and Information Systems Vellore Institute of Technology School of Computer Science Vellore India Sastra Deemed University School of Electrical & Electronics Engineering Thanjavur India

ISBN: (纸本)9798350353068

In today's digital world, safeguarding data at all times is important, especially with the widespread use of images across various processes. Image encryption is vital in concealing sensitive information through cryptographic techniques, rendering data indecipherable to unauthorized parties. However, existing encryption systems often fall short in terms of both efficiency and security. This research explores chaos theory for image encryption methodologies to address these limitations. A comprehensive study of image encryption fundamentals was conducted, including chaos theory, various chaotic maps, DNA encoding, and encryption system evaluation metrics. Implementation of different chaotic maps such as the logistic map, tent map, and combined logistic-tent map, along with 3D chaotic systems like the Lorenz and Rossler systems, was performed for encryption purposes. Additionally, the 4D extension of the Rossler hyperchaotic system was introduced for enhanced encryption capabilities. Encryption methods were evaluated using metrics such as UACI, NPCR, correlation coefficients, image entropy and histogram analysis, with results compared and tabulated. This research advances image encryption techniques by leveraging chaos theory to achieve heightened security and efficiency in safeguarding image data. © 2024 IEEE.

关键词： Chaotic systems

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：