检索结果-内蒙古大学图书馆

Reflective clothing detection based on YOLOv5s fused with SE attention mechanism

International Journal of Reasoning-based Intelligent Systems 2025年第1期17卷 9-16页

作者： Wang, Yuanyuan Gu, Dingyun Zhu, Yemeng Zhang, Zhihan Cao, Jiahui Zhang, Haiyan School of Computer and Software Engineering Huaiyin Institute of Technology Huaian223001 China Engineering Laboratory of Mobile Interconnection Technology for Internet of Things Jiangsu Province Huai’an Huaian City223001 China

This paper proposes a YOLOv5s deep learning algorithm incorporating the SE attention mechanism to address the issue of workers failing to wear reflective clothing on duty, which has resulted in casualties from time to time. The YOLOv5s model is used to train the reflective clothing dataset obtained by collecting images of construction sites, and the SE attention mechanism module is added to the network structure of YOLOv5 for improved performance. The reflective clothing detection algorithm is then deployed to the web for real-time detection. The experimental results show that the algorithm achieves mAP value of 96.70% with the SE attention module, resulting in an accuracy improvement of 1.09% compared to the model without the SE module. This enables practical applications to reduce the incidence of construction accidents at construction sites. Coрyright © 2025 Inderscience Enterрrises Ltd.

关键词： attention mechanism flask reflective clothing detection vue YOLOv5

来源：评论

学校读者我要写书评

暂无评论

HGLFNet: Hybrid Global Semantic and Local Detail Feature Network for Lane Detection

引用

IEEE Access 2025年 13卷 72314-72328页

作者： Ding, Lei Tang, Chunhui Fang, Yi University of Shanghai for Science and Technology School of Optical-Electrical and Computer Engineering Shanghai200093 China Shanghai Jiao Tong University School of Electronic Information and Electrical Engineering Shanghai200240 China

Lane detection is a crucial component of autonomous driving perception systems. Existing methods still encounter performance limitations in complex situations like occlusions and extreme illumination. To address these challenges, this paper introduces a novel Hybrid Global Semantic and Local Detail Feature Network (HGLFNet), designed to enhance lane detection accuracy and robustness in complex scenarios. Technically, we propose a Global Semantic Feature Extraction Module (GSFEM), integrating a Long-Range Information Capture Unit (LRICU) and a Multi-Scale Aggregation Unit (MSAU). GSFEM innovatively utilizes large-kernel convolutions to efficiently capture global contextual information, significantly improving the model’s perception of complete lane line shapes and overcoming the limited receptive field of conventional methods. Furthermore, we develop a Semantic Feature Fusion Module (SFFM) to effectively bridge the semantic gap between global semantic features and local detailed features. SFFM incorporates a Deformable Feature-based Attention Mechanism (DFAM) to adaptively enhance the representational power of fused features and highlight crucial lane line characteristics. HGLFNet effectively integrates global semantic context with local detailed information, enhancing the network’s ability to detect thin and occluded lane line structures. In quantitative evaluation, HGLFNet is compared with state-of-the-art methods. On the VIL-100 dataset, the F1 score and accuracy are improved by 0.75% and 0.38%, respectively, compared to ADNet. On the Tusimple dataset, the F1 score and accuracy are improved by 0.49% and 0.11%, respectively. On the more challenging CULane dataset, the F1 score is improved by 0.03%. The experimental results comprehensively demonstrate that HGLFNet surpasses the existing state-of-the-art techniques in both accuracy and efficiency, providing a novel and effective solution for lane detection in complex scenarios and showing significant potential in pra

关键词： Semantics Feature extraction Lane detection Shape Attention mechanisms Accuracy Kernel Lighting Autonomous vehicles Roads

来源：评论

学校读者我要写书评

暂无评论

Knowledge Probabilization in Ensemble Distillation: Improving Accuracy and Uncertainty Quantification for Object Detectors

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第1期6卷 221-233页

作者： Yang, Yang Wang, Chao Gong, Lei Wu, Min Chen, Zhenghua Li, Xiang Chen, Xianglan Zhou, Xuehai University of Science and Technology of China School of Computer Science and Technology Hefei230026 China 138632 Singapore East China Normal University School of Data Science and Engineering Shanghai200062 China

Ensemble object detectors have demonstrated remarkable effectiveness in enhancing prediction accuracy and uncertainty quantification. However, their widespread adoption is hindered by significant computational and storage demands, limiting their feasibility in resource-constrained settings. To overcome this, researchers have focused on distilling the knowledge from ensemble object detectors into a single model. In this article, we introduce probabilization based ensemble distillation (ProbED), an innovative ensemble distillation framework that consolidates knowledge from multiple object detectors into a single, resource-efficient model. Unlike traditional ensemble distillation methods that average the outputs of subteachers, ProbED captures comprehensive outcome distributions from all subteachers, providing a more nuanced approach to knowledge transfer. ProbED employs knowledge probabilization to achieve a sophisticated and refined aggregation of teacher knowledge, including feature knowledge, semantic knowledge, and localization knowledge, resulting in dual improvements in prediction accuracy and uncertainty quantification for the student model. In particular, ProED's novel knowledge probabilization-based approach to aggregating teacher knowledge is inspired by our empirical observations, which demonstrate that knowledge probabilization excels in effectively representing uncertainty, improving prediction, and facilitating robust knowledge transfer. Furthermore, we introduce a random smoothing perturbation technique to modify inputs within ProbED, further enhancing the distillation process. Extensive experiments highlight ProbED's ability to significantly enhance the prediction accuracy and uncertainty quantification of various object detectors, demonstrating its superior performance compared to other state-of-the-art techniques. © 2024 IEEE.

关键词： Teaching

来源：评论

学校读者我要写书评

暂无评论

Harnessing Light Field Angular Cues and Spatial Geometries for Semantic Segmentation

Harnessing Light Field Angular Cues and Spatial Geometries f...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Jia, Chen Shi, Fan Cheng, Xu School of Computer Science and Engineering Tianjin University of Technology Tianjin China

ISBN: (纸本)9798350368741

4D light field imaging captures rich spatial-angular information, providing essential geometric cues for semantic segmentation tasks. In this paper, we introduce a novel backbone network called the Light Field Extraction Interaction Network (LFEI-Net). LFEI-Net excels in extracting global structures and multi-scale spatial-angular features, capturing feature dependencies through channel modeling and diverse feature interactions. Unlike traditional methods that depend on pyramid and dilated feature extraction, LFEI-Net pioneers an efficient method by integrating large-scale horizontal depth-wise convolution (HDWC) and vertical depth-wise convolution (VDWC) with interactive operations for comprehensive spatial multi-scale feature extraction. Furthermore, we present the Multi-Angular Modeling (MAM) module, which effectively captures scene angle variations from multiple perspectives and precisely delineates object boundaries, thereby improving model adaptability. Our experimental evaluations on two datasets demonstrate that LFEI-Net significantly outperforms state-of-the-art (SOTA) 2D and 4D light field semantic segmentation methods, achieving mean Intersection over Union (mIoU) of 83.72% and 86.88%, respectively. © 2025 IEEE.

关键词： LFEI-Net light field imaging MAM module semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Optimized-Memory and Enhanced-Security Approaches to RSA and ElGamal

Optimized-Memory and Enhanced-Security Approaches to RSA and...

引用

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

作者： Kumar, Maheswer Sunil Balaji, Vinithra Sakkarapani, Sandosh School of computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331509828

This paper demonstrates 2 methods, a reduced memory technique, and a reduced memory along with more security techniques in RSA (Rivest-Shamir-Adleman) and ElGamal which are both asymmetric cryptographic algorithms. Regular RSA and ElGamal are known to be performance efficient, but memory wise, they were not efficient. Also, to enhance security of any algorithm, the key size is increased leading to more memory and computation. To solve these challenges, hashing method is used through which, the memory can be reduced without affecting the security or the performance of the algorithms. Along with these hashmaps, we create an array manipulation method, which enhances the security of data in the algorithm by making the cipher text more complex. The above is performed using pipe-lining and use of python. The objective of this research work is to optimize RSA and ElGamal algorithm with respect to memory load and increasing security, without affecting the overall computational overhead of the system. The results aim to contribute to the advancement of these cryptographic techniques and especially resource constrained environments, where memory is an important factor and for more security required procedures. The hashing method optimizes the performance of RSA and ElGamal and their use in transferring large confidential files and the array method enhances the security using a modulus acting as another key. Furthermore, this optimized technique can be applied to other cryptographic solutions © 2025 IEEE.

关键词： Pipe linings

来源：评论

学校读者我要写书评

暂无评论

Design of a Scoring System for National Fitness Volunteer Services under Deep Learning 4

Design of a Scoring System for National Fitness Volunteer Se...

引用

4th International Conference on Evolutionary Computing and Mobile Sustainable Networks, ICECMSN 2024

作者： Singh, Abhimeet Singla, Raghav Sodhi, Pulkit Pandian, M. Thurai School of Computer Science and Engineering Vellore Institute of Technology Vellore India

This work is developed to establish a comprehensive, scientific, and reasonable national fitness volunteer service scoring system under a long short-term memory (LSTM) recurrent neural network (RNN) algorithm. The LSTM RNN algorithm-based architecture of a three-layer information processing system for human motion recognition is proposed, which includes a data acquisition layer, a data calculation layer, and a data application layer. The LSTM RNN model is verified on the public dataset PAMAP2. Under the LSTM RNN-based information processing system for human motion recognition, the national fitness volunteer service scoring system is established by using the literature method, the survey method, the analytic hierarchy process (AHP) method, and the Delphi method. The candidate indicators are screened regarding the dimension of "service quality", the indicators are screened by the Delphi method, and the judgment matrix weights and consistency of each included indicator are analyzed. It is found that the information processing system for human motion recognition based on the LSTM RNN algorithm can effectively identify different motion states, such as sitting, lying down, running, walking, and cycling. A national fitness volunteer service scoring system based on the quality of communal sports facilities, services, fitness environment, affiliated sports facilities, and mass fitness service personnel is established under the LSTM RNN-based human motion recognition information processing system. The system includes one number of(#) first-level indicator, five number of(#) second-level indicators, and thirty two number of (#) third-level indicators. In summary, the LSTM RNN-based human motion recognition information processing system can correctly mine the types of fitness exercises, and the established national fitness volunteer service scoring system provides a scientific reference for building a perfect national fitness service system. © 2025 The Authors. Published by Els

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis of Financial News: Implications for Corporate Strategy in Risk Management, Mergers and Marketing

Sentiment Analysis of Financial News: Implications for Corpo...

引用

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

作者： Sendhil, R. Lijo, Joshua Abinaya, E. School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331509828

This research study examines the ways to use sentiment analysis on financial news for corporate strategy making. We examine the impact of sentiment in financial news on corporate decisions (beyond technology) regarding risk management, mergers and marketing strategy using recent developments from research into sentiment analysis as input and contrasts it with our findings. This study presents unsupervised and supervised sentiment analysis methods, statistical methods and AI tools for sentiment analysis, and their contributions to identifying business needs. In addition, the paper underscores existing difficulties and received a number of limitations on current techniques in sentiment analysis that have to be tended for better supporting corporate strategy. © 2025 IEEE.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

Secure File Sharing System Using Hybrid Authentication and Secured Key Management Algorithm 4th

Secure File Sharing System Using Hybrid Authentication and S...

引用

4th International Conference on Smart IoT Systems: Innovations and Computing, SSIC 2023

作者： Dokania, Aryan Bamb, Anshul Dheeba, J. School of Computer Science and Engineering Vellore Institute of Technology Vellore India

ISBN: (纸本)9789819751990

When web file sharing is used, users can safely receive and share any files or documents over the web using any of their preferred web browsers. Though the existing cryptographic algorithms perform well in end-to-end encryption, they are not resistant to attacks that extract information from the computer system, such as passwords and encryption/decryption keys. In this paper, a hybrid method is proposed for secure and efficient end-to-end encryption. This paper focuses on secure key exchange, efficient data encryption, and strengthening of passwords, which is achieved by a combination of four cryptographic algorithms. The proposed hybrid cryptographic method is secure against several attacks, including man-in-the-middle attacks and side-channel attacks. The proposed method offers many advantages over traditional end-to-end encryption methods, including security, efficiency, and resistance to side-channel attacks. Extensive analytical and experimental results show that the proposed method is efficient enough to be used in a file sharing system. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Authentication Protocol

来源：评论

学校读者我要写书评

暂无评论

Toward Collaborative Intelligence for Meta-Computing-Driven IIoT Based on Vertical Federated Learning With Fast Convergence

引用

IEEE Internet of Things Journal 2025年第10期12卷 13806-13816页

作者： Li, Youqi Liu, Shuangji Meng, Yanchen Qi, Shenyi Qu, Zhe Li, Fan Wang, Yu Beijing Institute of Technology School of Computer Science and Technology Beijing100081 China Central South University School of Computer Science and Engineering Changsha410083 China Temple University Department of Computer and Information Sciences PhiladelphiaPA19122 United States

Industrial Internet of Things (IIoT) is an emerging technology that digitizes industrial production and realizes Industry 4.0. However, it shows that IIoT is difficult to enable sophisticated downstream applications without eliciting all devices to achieve collaborative intelligence. Existing works on IIoT either require the consolidation of various IIoT devices’ data into a single centralized server which has potential privacy breach, or coordinate devices to learn a global model in privacy-preserving federated learning (FL) but assume data across devices has the sample feature space and neglect the heterogeneity of IIoT devices. In this article, we propose Meta-computing-driven vertical FL (VFL) algorithms to achieve collaborative intelligence in IIoT where heterogeneous devices have imperfect data with incomplete features. Specifically, we first provide the modeling of N devices’ VFL to collectively train the submodels and the common model. We present the computing graph to clearly indicate the gradient evaluation. To enable a fast convergence performance, we design a variance-reduced gradient estimator that can be seamlessly integrated into the basic VFL. Finally, we evaluate our proposed VFL by conducting experiments on the MNIST dataset regarding image recognition and the DAWM dataset for detecting anomalies in wafer manufacturing. The experimental results show that our VFL for IIoT is both effective and efficient. © 2014 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent hierarchical workflow for autonomous code generation with Large Language Models

Multi-Agent hierarchical workflow for autonomous code genera...

引用

2025 IEEE International Students' Conference on Electrical, Electronics and computer Science, SCEECS 2025

作者： Akilesh, S. Sekar, Rajeev Om Kumar, C.U. Prakalya, D. Suguna, M. Vellore Institute of Technology School of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331529833

This paper presents a multi-agent hierarchical workflow tailored for automating data analysis, code generation, and visualization, focusing specifically on user-provided CSV datasets. The workflow integrates AlphaCodium with LangChain, LangGraph, and GPT, enabling autonomous code generation, unit test development, and debugging. The system operates by analyzing the uploaded dataset, assigning agents that work sequentially: a programmer agent generates code using the Skeleton-Of-Code approach, a unit test agent verifies the code, and an executor agent runs the code. A debugging agent is also included to identify and resolve any issues. The AI agents involved are capable of dynamically accessing online resources, including documentation and references to enhance their decision-making processes. This work attempts to exemplify the application of AI in automating not only code execution but also the planning and validation stages by writing unit tests in the software development process to reflect the increasing role of AI in advancing automation for rapid code generation within the software industry. © 2025 IEEE.

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：