检索结果-内蒙古大学图书馆

Tackling Real-world Complexity: Hierarchical Modeling and Dynamic Prompting for Multimodal Long Document Classification

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年第6期35卷 5776-5790页

作者： Liu, Tengfei Hu, Yongli Li, Mingjie Yi, Junfei Chang, Xiaojun Gao, Junbin Yin, Baocai Beijing University of Technology Beijing Key Laboratory of Multimedia and Intelligent Software Technology Beijing Institute of Artificial Intelligence School of Information Science and Technology Beijing100124 China Stanford University Department of Radiation Oncology StanfordCA94305 United States Hunan University National Engineering Laboratory of Robot Visual Perception and Control Technology College of Electrical and Information Engineering Hunan Changsha410082 China University of Technology Sydney ReLER Laboratory Australian Artificial Intelligence Institute Faculty of Engineering and Information Technology UltimoNSW2007 Australia The University of Sydney Business School The University of Sydney Discipline of Business Analytics CamperdownNSW2006 Australia

With the rapid growth of internet content, multimodal long document data has become increasingly prominent, drawing significant attention from researchers. However, most existing methods primarily focus on scenarios where all modalities are present, often overlooking more challenging and realistic cases involving missing image modality. To address this limitation, we propose a robust multimodal long document classification (MLDC) framework that integrates hierarchical modeling and dynamic prompting to handle complex multimodal long document data. Our approach begins by leveraging hierarchical modeling combined with an Adaptive Correlation Multimodal Transformer (ACMT) to effectively capture relationships between text and images at both section and sentence levels. We also introduce a Dynamic Prompt Generation (DPG) module at both levels to enhance the model's robustness in handling missing image data. By evaluating sample uncertainty, the DPG module dynamically adjusts both the number of prompts and the prompts themselves, allowing the model to better adapt to the varying needs of different samples. Finally, a Hierarchical Heterogeneous Graph (HHG) is introduced to enhance feature interactions across levels, further improving the coherence and accuracy of the model. Extensive experiments on four multi-modal long document datasets demonstrate that our model shows superior performance compared to existing state-of-the-art MLDC classification methods in various conditions. © 1991-2012 IEEE.

关键词： Adaptation models Transformers Data models Correlation Circuits and systems Electronic mail Robustness Uncertainty Training Complexity theory

来源：评论

学校读者我要写书评

暂无评论

A Novel Conv-Mamba-Hybrid Network for Medical Image Segmentation

A Novel Conv-Mamba-Hybrid Network for Medical Image Segmenta...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Qiao, Sibo Zhao, Zhiyuan Xie, Pengfei Yin, Wenjing Wang, Min Zhang, Yuanyuan Pang, Shanchen Tiangong University The College of Software Tianjin China China University of Petroleum The College of Computer Science and Technology Qingdao China Khalifa University of Science and Technology The Department of Electrical Engineering and Computer Science Abu Dhabi United Arab Emirates Tiangong University The College of Life Sciences Tianjin China Qingdao University of Technology The College of Information and Control Engineering Qingdao China

ISBN: (纸本)9798350386226

In the rapidly advancing landscape of intelligent medical technology, high-precision and widely applicable medical image segmentation techniques are pivotal in advancing personalized treatment plans and enhancing patient experiences. To address inherent obstacles, such as complex structures and features that are difficult to capture in segmentation tasks, we first propose a CMH module that merges convolutional neural networks (CNNs) with state space models (e.g., Mamba). This module hierarchically integrates CNN's local priors into the Mamba layer, adeptly capturing specific details of target objects and broader contextual information from different perspectives. Subsequently, we introduce an information fusion module based on Mamba (MIH), aimed at dynamically fusing feature information from different paths and levels to enhance feature expression and discrimination. Building upon the abovementioned modules, we develop an unexplored segmentation model, CMHNet, capable of flexibly capturing feature maps of varying scales and angles from medical images. In the experiments, we present CMHNet's distinguished performance in segmentation tasks through massive experiments on four medical datasets. © 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

Pretrained Model with CNN for Lung Cancer Identification and Classification 7

Pretrained Model with CNN for Lung Cancer Identification and...

引用

7th International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2024

作者： Kanwal, Shaista Ali, Sajid Alahmadi, Tahani Jaser Assam, Muhammad Ansari, Ghulam Jillani Bhatti, Uzair Aslam University of Education Department of Information Sciences Lahore Pakistan College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University Department of Information Systems P.O. Box 84428 Riyadh Saudi Arabia University of Science and Technology Department of Software Engineering KPK Bannu Pakistan School of Information and Communication Engineering Hainan University China

ISBN: (纸本)9798350350890

Lung cancer impacts both genders, and its early detection is key to lowering death rates. Current deep learning techniques for automated identification and classification of lung carcinoma face issues with interpretability, data variability handling, and broader applicability. In our research, we propose a novel method that merges a convolutional neural network (CNN) with a streamlined RESNET18 framework. This approach is designed for lung cancer detection and categorization from CT scans, involving image preprocessing like resizing and rescaling. It effectively differentiates between healthy and cancerous tissues, further classifying them as either Normal, Benign, or Malignant, including variants such as adenocarcinoma and squamous cell carcinoma. Our CNN+RESNET18 model boasts an impressive 99% accuracy, supporting radiologists in the intricate task of image analysis, reducing risks of incorrect or delayed diagnosis, and fostering prompt treatment for improved patient outcomes. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Rail Surface Defect Detection Method 9

Unsupervised Rail Surface Defect Detection Method

引用

9th International Conference on Intelligent Informatics and BioMedical sciences, ICIIBMS 2024

作者： Zhao, ChenYu Han, Yi Ouyang, Ningkang Zhao, YiFan School of Artificial Intelligence Zhengzhou Software Vocational and Technical College Zhengzhou China Anyang Institute of Technology Engineering Department of Computer Science and Information Anyang China North China University of Water Resources and Electric Power School of Information Engineering Zhengzhou China

ISBN: (纸本)9798350363043

Railroad travel has become an essential part of modern life, therefore, rail surface defect detection has become a problem that cannot be ignored. However, most of the currently existing methods often require a large number of defect samples for learning, which are labor-intensive and very difficult to obtain. In order to alleviate the data dependence of the detection model, this paper proposes an unsupervised method. Specifically, we use Transformer's attention mechanism to construct the encoder-decoder structure and train it only on normal samples with the purpose of reconstruction, so that those regions that cannot be reconstructed will be regarded as defects at the time of detection. To prevent the model from generalizing its generalization ability to those unknown defective samples, i.e., to avoid overgeneralization, we introduce a noise perturbation in the training so that the model focuses on the reconstruction of normal features. Experiments show that the proposed unsupervised approach outperforms a host of supervised models with low data requirements and can effectively address the current challenges of track defect detection. © 2024 IEEE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

The Way to Check the Reliability Test in Smart Grids for the Smart Protection System

The Way to Check the Reliability Test in Smart Grids for the...

引用

2023 International Conference on Power Energy, Environment and Intelligent Control, PEEIC 2023

作者： Hannah Jessie Rani, R. Goyal, Manish Kumar Mishra, Awakash Faculty of Engineering and Technology Department of Electrical and Elceltronics Engineering Bangalore India Vivekananda Global University Department of Computer Science & Application Jaipur India Maharishi University of Information Technology Engineering Department Uttar Pradesh India

ISBN: (纸本)9798350357769

The complex operation of contemporary power systems is closely linked to the widespread integration of computer networks, or cyber networks. These applications fall into two general categories: direct and indirect interdependencies, which are the result of how they communicate with power networks. Past studies explored the area of direct interdependencies in cyber-power networks and presented a state mapping-based model that was carefully designed to assess its impact on power system dependability. The investigation of indirect relationships between cyber and electricity networks is the new emphasis of this article. A wide range of applications involving indirect interdependencies in modern power systems are examined, laying the foundation for the release of a model that is based on state updating. This model is carefully crafted to provide a numerical evaluation of cyber-power networks' dependability in the face of indirect interdependencies. Two separate models are used to maximize data communication inside the cyber network and minimize load limitation in the power network, respectively, in order to maximize the functioning of these integrated networks. In order to demonstrate the efficacy of the suggested solution model, the research focuses on a high-voltage converter equipped with both protection and monitoring devices. By taking a comprehensive approach, the research makes a substantial contribution to a deeper understanding of the complex interactions between cyber and electricity networks, providing priceless insights for enhancing the dependability of these interconnected systems. © 2023 IEEE.

关键词： Cyber-Power Interdependencies Dependability Indices and Optimization Framework Indirect Relationships Monitoring Systems P- Table Protection Systems Reliability Assessment Smart Grid State Updating

来源：评论

学校读者我要写书评

暂无评论

Automatic Recognition of Construction Worker Activities Using Deep Learning Approaches and Wearable Inertial Sensors

引用

Intelligent Automation & Soft Computing 2023年第5期36卷 2111-2128页

作者： Sakorn Mekruksavanich Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication TechnologyUniversity of PhayaoPhayao56000Thailand Department of Mathematics Intelligent and Nonlinear Dynamic Innovations Research CenterFaculty of Applied ScienceKing Mongkut’s University of Technology North BangkokBangkok10800Thailand

The automated evaluation and analysis of employee behavior in an Industry 4.0-compliant manufacturingﬁrm are vital for the rapid and accurate diagnosis of work performance,particularly during the training of a new *** techniques for identifying and detecting worker performance in industrial applications are based on computer vision *** widespread com-puter vision-based approaches,it is challenging to develop technologies that assist the automated monitoring of worker actions at external working sites where cam-era deployment is *** the use of wearable inertial sensors,we propose a deep learning method for automatically recognizing the activities of construction *** suggested method incorporates a convolutional neural network,residual connection blocks,and multi-branch aggregate transformation modules for high-performance recognition of complicated activities such as con-struction worker *** proposed approach has been evaluated using standard performance measures,such as precision,F1-score,and AUC,using a publicly available benchmark dataset known as VTT-ConIoT,which contains genuine con-struction work *** addition,standard deep learning models(CNNs,RNNs,and hybrid models)were developed in different empirical circumstances to compare them to the proposed *** an average accuracy of 99.71%and an average F1-score of 99.71%,the experimentalﬁndings revealed that the suggested model could accurately recognize the actions of construction ***,we examined the impact of window size and sensor position on the identiﬁcation efﬁciency of the proposed method.

关键词： Complex human activity recognition wearable inertial sensors deep learning construction workers automatic recognition

来源：评论

学校读者我要写书评

暂无评论

End-to-End Congestion Avoidance Scheme for Underwater High Speed Communication Networks 3

End-to-End Congestion Avoidance Scheme for Underwater High S...

引用

3rd International Conference on Smart Generation Computing, Communication and Networking, SMART GENCON 2023

作者： Shivank Jain, Romil Reena, R. Kadam, Prema S. Das, Sunanda Singh, Swati Vivekananda Global University Department of Computer Science & Engineering Jaipur India Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Chennai127 India Vishwakarma Institute of Information Technology Department of Artificial Intelligence & Data Science Pune India Computer Science and Engineering Faculty of Engineering and Technology Karnataka Bangalore India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (纸本)9798350319125

The cease-to-give up Congestion Avoidance Scheme for Underwater high pace verbal exchange Networks, commonly known as E2E-UCAN, is a critical technique for lowering plausible throughput by using efficiently warding off congestion inside a community. E2E-UCAN is a disbursed congestion control mechanism which uses statistics obtained from cease-to-end feedbacks to regulate sending prices on the give up systems. By means of monitoring the differences among expected network overall performance and actual performance, E2E-UCAN is able to pick out congestion stages inside the device, and is therefore able to regulate to prevent congestion before it happens. This enables to improve the stableness of the community and to increase the throughput of a network. Moreover, E2E-UCAN is likewise capable of efficaciously coping with queuing, avoiding packet loss in non-uniform networks, and in preferred, permitting applications to maximize performance. © 2023 IEEE.

关键词： Community Congestion Lowering Mechanism Performance

来源：评论

学校读者我要写书评

暂无评论

Achieving Efficient and Accurate Segmentation for Autonomous Driving 9

Achieving Efficient and Accurate Segmentation for Autonomous...

引用

9th International Conference on Intelligent Informatics and BioMedical sciences, ICIIBMS 2024

作者： Ouyang, Ningkang Dong, Jie Han, Yi Zhao, ChenYu North China University of Water Resources and Electric Power School of Information Engineering Zhengzhou China Anyang Institute of Technology Engineering Department of Computer Science and Information Anyang China Zhengzhou Software Vocational and Technical College School of Artificial Intelligence Zhengzhou China

ISBN: (纸本)9798350363043

Autonomous driving technology, as a mainstream trend in today's technological development, holds significant commercial value. Semantic segmentation, a core technology in this field, faces challenges with current mainstream models due to their complex structures, large numbers of parameters, and the need for high-performance hardware. These issues hinder their application in mobile devices. Therefore, this paper proposes a road scene semantic segmentation algorithm based on dynamic asymmetric convolution and capsule networks, named Asymmetric Capsule Convolutional Networks (ACCN). Specifically, the algorithm uses dynamic asymmetric convolution to construct a baseline network that combines dynamic convolution and capsule dynamic routing algorithms to enhance the model's representational capabilities. Additionally, a dispersed attention module is built to strengthen the dependency between features to achieve more adaptable network weights. Experiments demonstrate that ACCN achieves a good balance between segmentation accuracy and detection speed. On the CamVid and Cityscapes test datasets, the proposed method in this paper achieved an average Intersection over Union (mIoU) of 80.8% at 55.2 fps, showcasing better trade-offs in segmentation accuracy and efficiency. The experimental results indicate the effectiveness of our method. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Exploring the Use of Machine Learning Techniques for Enhancing Real-Time Data Access 3

Exploring the Use of Machine Learning Techniques for Enhanci...

引用

3rd International Conference on Smart Generation Computing, Communication and Networking, SMART GENCON 2023

作者： Rampal, Sourav Shreenidhi, H.S. Shrivastava, Manish Singh, Akhilendra Pratap Reena, R. Kumar, Santosh D. Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Computer Science and Engineering Faculty of Engineering and Technology Karnataka Bangalore India Vivekananda Global University Department of Computer Science & Engineering Jaipur India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Chennai127 India Vishwakarma Institute of Information Technology Department of Artificial Intelligence & Data Science Pune India

ISBN: (纸本)9798350319125

Actual-time statistics to date has updated, increasingly vital as the quantity of virtual information maintains updated growth. Effective choice-making depends on the capability to update access and analyze records. Device up-to-date techniques provide the capacity up-to-date enhance actual-time statistics up-to-date by using offering insights from disparate pieces of information. This paper addresses the potential benefits of using a system with date knowledge and updated, up-to-date strategies to broaden extra clever statistics up to date updated systems. It summarizes the contemporary nation of the art and highlights studies in machine learning and up-to-date imaginative and prescient real-time information up-to-date. It additionally discusses present-day challenges and shows viable instructions for future studies. The studies network up-to-date recall how gadgets are up to date know that updated strategies may be used to facilitate real-time get entry updated the maximum precious pieces of facts. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

TEXCEL: text encryption with elliptic curve cryptography for enhanced security

引用

Multimedia Tools and Applications 2025年第13期84卷 11503-11531页

作者： Sharma, P.L. Gupta, Shalini Monga, Himanshu Nayyar, Anand Gupta, Kritika Sharma, Arun Kumar Department of Mathematics & Statistics Himachal Pradesh University Shimla171005 India Government Hydro Engineering College Bilaspur174001 India Graduate School Faculty of Information Technology Duy Tan University Da Nang550000 Viet Nam Department of Computer Science and Engineering Central University of Jammu District Samba J&K 181143 India

Elliptic curve cryptography has emerged as a potent tool for safeguarding data communications. This research paper addresses the problem of traditional text encryption methods relying on shared code tables between senders and receivers. This reliance not only adds complexity but also increases the risk of security breaches and requires significant communication overhead. In the present paper, we introduce TEXCEL, an innovative, efficient and authenticated text encryption scheme. TEXCEL begins with a pioneering method for crafting a customized permuted ASCII table elevating cryptographic security by leveraging elliptic curve coordinates as keys. These coordinates serve as seeds for a random number generator ensuring reproducibility. Here, the technique used to map the characters of the message to a point on elliptic curve eliminates the traditional approach of sharing a code table between sender and receiver, thus reducing communication overhead. TEXCEL undergoes a thorough security analysis, including NIST evaluation, key and plaintext sensitivity, and resilience against common attacks like known-ciphertext, known-plaintext, chosen-plaintext, and chosen-ciphertext. To test and validate the proposed methodology, experimental results present that TEXCEL demonstrates plaintext sensitivity exceeding 98%, high randomness, and a robust key space. Additionally, encryption and decryption of a 5000-character text takes only 0.027 s and 0.007 s respectively, with a cipher size of 9.8 KB, ensuring both security and efficiency. Furthermore, as compared with the existing schemes, TEXCEL shows enhanced resilience against modern cryptanalysis and outperforms existing schemes with a 53.50% improvement in encryption time, 94.86% in decryption time, 53.31% in cipher data size and better plaintext sensitivity. These findings justify the suitability of TEXCEL for secure text communication in diverse real-time applications. © The Author(s), under exclusive licence to Springer science+Bu

关键词： Geometry

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：