检索结果-内蒙古大学图书馆

Hybrid CNN-ViT architecture to exploit spatio-temporal feature for fire recognition trained through transfer learning

引用

Multimedia Tools and Applications 2025年第8期84卷 4703-4732页

作者： Shahid, Mohammad Wang, Hong-Cyuan Chen, Yung-Yao Hua, Kai-Lung Dept. of Computer Science and Information Engineering National Taiwan University of Science and Technology Taipei Taiwan Dept. of Electronic and Computer Engineering National Taiwan University of Science and Technology Taipei Taiwan

Fires are becoming one of the major natural hazards that threaten the ecology, economy, human life and even more worldwide. Therefore, early fire detection systems are crucial to prevent fires from spreading out of control and causing destruction. Based on vision sensors, many fire detection techniques have evolved with the recent surge of curiosity in deep learning, which exploits the spatial features of individual images. However, fire can take different forms, scales, and combustion materials can produce different colors, making accurate fire detection from an image challenging. Small fires captured from long-distance cameras lack salient features, further complicating detection. This paper proposes a hybrid structure that uses attention-enhanced convolutional neural networks and vision transformers (CNN-ViT) to detect fire. The proposed CNN-ViT first pays spatial attention to every frame and then aggregates temporal contextual information from neighboring frames to improve detection performance. Due to the limited availability of training fire datasets, the study employs deep transfer learning for feature extraction using pre-trained CNN. We used various metrics to examine the efficacy of the proposed approach. The results showed that the CNN-ViT method outperformed previous models based on spatial-temporal characteristics by achieving a relative improvement in accuracy and F1 score. The satisfactory results on images contaminated with different intensities of noise confirm the robustness of the approach. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

An Intelligent LLM-Powered Personalized Assistant for Digital Banking Using LangGraph and Chain of Thoughts 22

An Intelligent LLM-Powered Personalized Assistant for Digita...

引用

22nd IEEE International Symposium on Intelligent Systems and Informatics, SISY 2024

作者： Easin, Arafat Md Sourav, Saha Tamas, Orosz Data Science and Engineering Dept. Budapest Hungary Dept. of Computer Science and Engineering Dhaka Bangladesh

ISBN: (纸本)9798350385601

Text-based applications and chatbots are increasingly popular for delivering banking services and educational tools, offering convenient and efficient solutions for users. Whereas, personalized assistants have transformed user engagement in the digital banking space by utilizing Large Language Models (LLMs) in conjunction with autonomous agents. This study proposes the development of an intelligent personalized assistant for digital banking, utilizing a multi-agent framework based on the LangGraph and Chain of Thoughts (COT) prompting. While COT guarantees context-aware replies, the LangGraph design maps characteristics to nodes to improve user interactions. The objectives of this system are to enhance task efficiency and elevate the capabilities of digital banking assistants. We present a customizable digital banking system powered by LLM-based models, designed to deliver an interactive and personalized banking experience. The system supports a range of services, including adding money, transferring funds, paying bills, accessing telco services like mobile recharge, managing savings interest rates, DPS schemes, fixed deposits, and answering FAQs related to banking information. Therefore, integrating COT for logical reasoning enhances the effectiveness of multi-agent systems, as each single agent benefits from the structured reasoning process. In addition, LangGraph is employed for structured data management, enabling the assistant to support and accelerate various digital banking processes efficiently. The code implementation of this work is available for public access at: https://***/srv-sh/digital-agent. © 2024 IEEE.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

MOPO-HBT: A movie poster dataset for title extraction and recognition

引用

Multimedia Tools and Applications 2024年第18期83卷 54545-54568页

作者： Ghosh, Mridul Roy, Sayan Saha Banik, Bivan Mukherjee, Himadri Obaidullah, Sk Md Roy, Kaushik Dept. of Computer Science Shyampur Siddheswari Mahavidyalaya Howrah India Dept. of Computer Science & Engineering Aliah University Kolkata India Dept. of Electronics & Electrical Communication Engineering IIT Kharagpur Kharagpur India TISA Lab Dept. of Computer Science West Bengal State University Barasat India

Real-world images often encompass embedded texts that adhere to disparate disciplines like business, education, and amusement, to name a few. Such images are graphically rich in terms of font attributes, color distribution, foreground-background similarity, and component organization. This aggravates the difficulty of recognizing texts from these images. Such characteristics are very prominent in the case of movie posters. One of the first pieces of information on movie posters is the title. Automatic recognition of movie titles from images can aid in efficient indexing as well as information conveyance. However, it is accompanied by other texts like the names of actors, producers, taglines, dates, etc. Though the organization of components is somewhat similar across different film industries like Tollywood (West Bengal), Bollywood (Mumbai), and Hollywood (Los Angeles), the graffiti patterns differ in multifarious instances. To address the problem of movie title understanding, we propose a dataset named MOvie POsters-Hollywood Bollywood Tollywood (MOPO-HBT) that encompasses movie posters from the aforementioned industries. The entire dataset is publicly available (http://***/11564) for research purposes. The baseline results of title identification and recognition were obtained with a CNN-based (Convolutional Neural Network) approach, wherein the titles were extracted using the M-EAST (Modified-Efficient and Accurate Scene Text) detector model. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

J-NECORA: A Framework for Optimal Resource Allocation in Cloud-Edge-Things Continuum for Industrial Applications With Mobile Nodes

引用

IEEE Internet of Things Journal 2025年第11期12卷 16525-16542页

作者： Pettorali, Marco Righetti, Francesca Vallati, Carlo Das, Sajal K. Anastasi, Giuseppe Univ. of Pisa Dept. of Information Engineering Italy Missouri Univ. of Science and Technology Dept. of Computer Science RollaMO United States

In the Industrial Internet of Things (IIoT) landscape, where the Cloud-to-Things Continuum (C2TC) paradigm is now a reality, industrial applications need to cope with highly heterogeneous network and computing resources. Moreover, many industrial applications also involve Mobile Nodes (MNs). Efficient allocation of network and computing resources to meet the stringent requirements of such applications is often a very challenging task. In this paper, we propose J-NECORA (Joint NEtwork and COmputing Resource Allocation), a comprehensive analytical framework to derive the optimal joint allocation of network and computing resources in the C2TC, that guarantees the application requirements, even in the presence of MNs. If the optimal allocation does not exist, J-NECORA provides a best-effort allocation that minimizes the delay. The proposed framework is highly modular and flexible, and can be easily customized to different scenarios. We validated its effectiveness through simulation experiments, focusing on a specific use case. For the considered scenarios, the performance predicted by our framework matches quite closely the simulation results. © 2014 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

AI in CS Education: Opportunities, Challenges, and Pitfalls to Avoid

引用

ACM Inroads 2024年第3期15卷 52-57页

作者： Vahid, Frank Dept. of Computer Science and Engineering University of California Riverside United States

Csinstructors realize education may be impacted by generative AI (artificial intelligence). This article describes (1) opportunities, like 24/7 help or auto-grading, (2) challenges, like increased cheating or student isolation, and (3) pitfalls to avoid, like inequitable access or false cheating accusations. The descriptions come from 15 CS educators and practitioners, from diverse institutions, who met at a 2023 NSF workshop on AI in CS education. The article aims to assist the CS community to take advantage of generative AI to improve courses, embark on new research, treat students equitably, and ultimately improve CS education for instructors and students. © 2024 Association for Computing Machinery. All rights reserved.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

TCCL: Discovering Better Communication Paths for PCIe GPU Clusters 24

TCCL: Discovering Better Communication Paths for PCIe GPU Cl...

引用

29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2024

作者： Kim, Heehoon Ryu, Junyeol Lee, Jaejin Dept. of Computer Science and Engineering Seoul National University Seoul Korea Republic of Dept. of Data Science Dept. of Computer Science and Engineering Seoul National University Seoul Korea Republic of

ISBN: (纸本)9798400703867

Exploiting parallelism to train deep learning models requires GPUs to cooperate through collective communication primitives. While systems like DGX, equipped with proprietary interconnects, have been extensively studied, the systems where GPUs mainly communicate through PCIe have received limited attention. This paper introduces TCCL, a collective communication library designed explicitly for such systems. TCCL has three components: a profiler for multitransfer performance measurement, a pathfinder to discover optimal communication paths, and a modified runtime of NCCL to utilize the identified paths. The focus is on ring-based collective communication algorithms that apply to popular communication operations in deep learning, such as AllReduce and AllGather. The evaluation results of TCCL on three different PCIe-dependent GPU clusters show that TCCL outperforms (up to ×2.07) the state-of-the-art communication libraries, NCCL and MSCCL. We also evaluate TCCL with DL training workloads with various combinations of parallelism types. © 2024 Copyright held by the owner/author(s).

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Moth flame optimization algorithm for efficient feature selection and hybrid classification technique for intrusion detection system

引用

Multimedia Tools and Applications 2025年 1-21页

作者： Keerthanarani, U. Selvakumar, V. Singaravelan, S. Raja, S. Edwin Thenmozhi, K. Shunmugam, D. Arun Dept. of Computer Science and Engineering P.S.R Engineering College Sivakasi India Dept.Of Computer Science and Engineering Vel Tech Rangarajan Dr Sagunthala R & d Institute of Science and Technology Chennai Avadi India Dept. Of Computer Science and Engineering Coimbatore Institute of Engineering and Technology Coimbatore India Dept. Of Electronics and Communication Engineering P.S.R Engineering College Sivakasi India

The rapid expansion of the Internet of Things (IoT) brings numerous benefits but further presents fresh difficulties, especially in terms of security. The distributed and interconnected nature of IoT devices makes them susceptible to various security threats. An Intrusion Detection System (IDS) plays a crucial role in safeguarding IoT networks by monitoring, detecting, and responding to potential security incidents. The Internet of Things (IoT) connects billions of devices, sensors, and systems, transforming industries and everyday life. However, this extensive interconnectivity also introduces significant security risks, making Intrusion Detection Systems (IDS) a critical component for safeguarding IoT environments. Feature selection is a crucial aspect of building effective Intrusion Detection Systems (IDS). Selecting relevant features not only helps in achieving better classification accuracy but also contributes to reducing false alarm rates. In the proposed system a moth flame optimizer is utilized to select the efficient features from the dataset, overall accuracy of over 97%. Further it is classified employing Random Forest and J48 classifier which attains high classification accuracy and low false alarm rate. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Differential Privacy-Based data Collection for Improving data Utility and Reducing Computational Overhead

引用

Transactions of the Korean Institute of Electrical Engineers 2025年第1期74卷 102-108页

作者： Kim, Jong Wook Dept. of Computer Science Sangmyung University Korea Republic of

With the rise of real-time data collection through mobile devices such as smartphones, user-driven decision-making systems in various fields such as transportation and healthcare have advanced significantly. However, increasing concerns about the privacy of sensitive information have emerged, leading to the widespread use of differential privacy (DP) to mitigate these risks. Among DP-based data collection methods, local differential privacy (LDP) and distributed differential privacy (DDP) are prominent. While LDP provides low computational overhead, it compromises data utility, while DDP provides better utility at the cost of higher computational overhead. To address these limitations, this paper introduces a DP-compliant (θ,λ) data collection method. The proposed approach collects a fraction of θ of the total items and a fraction of λ of the total user data, achieving improved data utility compared to LDP while significantly reducing computational overhead compared to DDP, resulting in more efficient data collection. Experimental results on real-world datasets show that the proposed method not only improves data utility over LDP, but also effectively reduces computational overhead over DDP. Copyright © The Korean Institute of Electrical Engineers.

关键词： data collection

来源：评论

学校读者我要写书评

暂无评论

Detecting Feet's Bending Patterns With Smart Insoles 5

Detecting Feet's Bending Patterns With Smart Insoles

引用

5th IEEE International Conference on Pattern Recognition and Machine Learning, PRML 2024

作者： Zhong, Yuchen Ying, Qijun Cai, Xiaohui dept. of Computer Science and Technology Hefei China dept. of Data Science Hefei China

ISBN: (纸本)9798350355925

Foot bending is a basic component of walking. Although bending is important for foot-ground interaction and ensures the speed and balance of the whole walking cycle, there is still no practical method for recording and analyzing foot bending under everyday scenarios. In this study we propose to use a smart insole, with two embedded Inertial Measurement Units, to acquire the movements of the fore-foot and the hind-foot and deduce the bending, represented by the bending point's position. We use the deep learning approach to estimate the bending position and its changes during different walking patterns. On the dataset containing 8 test subjects and 6 walking patterns, our method can achieve an overall mean absolute error accuracy of 2.58 cm. The smart insole can be with the subject all the day, bring in a new method for continuous foot bending monitoring. © 2024 IEEE.

关键词： Wearable technology

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence and Machine Learning Algorithms to Analyse Integrated Nutrient Management Strategies for Optimal Yield and Quality of Oilseeds Crops for Sustainable Development 4

Artificial Intelligence and Machine Learning Algorithms to A...

引用

4th International Conference on Soft Computing for Security Applications, ICSCSA 2024

作者： Mithra, C. Dar, Showkat A. Preethi, A. Dept. of Computer Science and Engineering Chennai India Githam University Dept. of Computer Science and Engineeing India Dept. of Computer Science and Engineering Chidambaram India

ISBN: (纸本)9798331515720

The main objective of this research is to deduce the efficacy of integrated nutrient management (INM) technologies in production of oilseed crops for sustainable development. A great amount of experience is needed in selecting the most effective INM strategy. A new recommendation system to circumnavigate this issue is proposed. It lets farmers decide on the best INM strategy to maximize oilseed crop yield and quality. This system is built on the techniques of advanced machine Learning (ML) and aritifical Intelligence (AI). Oilseed crop date in Tamil Nadu from 1961 to 2019 was used to develop the proposed algorithm. The proposed algorithm for crop yield prediction (CYP) which includes a Soft Voting Ensemble Classifier with weights (SVECWW), a Soft Voting Ensemble Classifier without weights (SVECWOW) along with the SVM technique are compared and contrasted with existing algorithms and also proves that SVECWW outperforms other ML algorithms with an accuracy rate of 97.2%. Furthermore, the Stacked Generalization Ensemble model is employed and compared with another Deep Neural Network (DNN) for the INM crop recommendation system which offers a simple graphical user interface (GUI) for farmers to use and received an accuracy of 97.5%. This GUI enables farmers to access valuable information such as the optimal timing for cultivating oilseed crops, the appropriate types and quantities of organic manures, inorganic fertilizers, and bio-fertilizers required for successful oilseed crop production. The study shows, on its whole, how to create tailored recommendation systems for farmers using GUI models with artificial intelligence and machine learning algorithms. Implementing these systems is expected to significantly improve oilseed crop production and quality significantly, benefiting the whole agricultural sector for sustainable development. Artificial intelligence (AI) makes a recommendation system more accurate and adaptable by looking through complex datasets and patterns

关键词： Adversarial machine 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：