检索结果-内蒙古大学图书馆

Attention-Enhanced Deep Reinforcement Learning for Electric Vehicle Routing Optimization

IEEE Transactions on Transportation Electrification 2025年

作者： Wang, Chao Zhang, Renyuan Hong, Rentao Wang, Haibo Anhui University State Key Laboratory of Opto-Electronic Information Acquisition and Protection Technology Engineering Research Center of Autonomous Unmanned System Technology Ministry of Education School of Artificial Intelligence Hefei230601 China University of Science and Technology of China Institute of Artificial Intelligence Hefei Comprehensive National Science Center School of Computer Science and Technology Hefei230601 China

Recently, the electric vehicle routing problem (EVRP) has emerged as a significant challenge in the transportation industry. Unlike traditional vehicle routing problems (VRPs) that focus primarily on route optimization for customers, the EVRP must also consider electricity constraints and charging decisions. Although deep reinforcement learning (DRL), particularly with attention-based Transformer architecture, has achieved success in VRPs, it faces significant difficulties when directly applied to the EVRP due to the additional multifaceted constraints and node types. To address these challenges, this paper proposes an attention-enhanced DRL method specifically designed for the EVRP. First, a heterogeneous attention mechanism is developed in the encoder to capture the intrinsic access relationships between customers and charging stations. Then, an embedding enhancement module is integrated to enrich node embeddings by incorporating adjacent distance and solution structure information through a global-local attention mechanism. Finally, a problem-size sensor is incorporated within the decoder to quantify route planning progress, thereby enhancing adaptability to the EVRP constraints. Empirical results demonstrate that the proposed method not only achieves higher efficiency compared to heuristic methods, but also significantly outperforms representative DRL-based approaches in terms of solution quality. © 2025 IEEE.

关键词： Vehicle routing

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Model based Face Mask Detection for Automated Mandation

Deep Learning Model based Face Mask Detection for Automated ...

引用

Advancement in Computation & computer Technologies (InCACCT), International Conference on

作者： Pratham Lokhande Shivangi Surati Himani Trivedi Bela Shrimali Information and Communication Technology Department Dhirubhai Ambani Institute of Information and Communication Technology Gandhinagar Gujarat India Department of Computer Science and Engineering School of Technology Pandit Deendayal Energy University Gandhinagar Gujarat India Department of Computer Engineering LDRP Institute of Technology and Research Kadi Sarva Vishwavidyalaya University Gandhinagar Gujarat India Department of Computer Science and Engineering Institute of Technology Nirma University Ahmedabad Gujarat India

SARS COVID19- virus has been a serious threat to human life, the best way to protect oneself is by wearing a face mask. As per World Health organization (WHO), wearing a mask can cut down the risk of infection by 85%. This study sought to offer an automated solution using several state-of-art Deep Convolution Neural Network (CNN) face recognition models to classify an individual’s face is wearing a mask or not. Determining the same and regulating the same is a difficult task that involves great human efforts. Nine various CNN models are trained, thereafter tested and found that the architecture of NasNet outperforms the other eight state-of-the-art models with respect to accuracy, on both, Real-World Masked Face Dataset (RMFD) and Simulated Masked Face Dataset (SMFD). The proposed approach achieved 100% testing accuracy on SMFD and 99.872% testing accuracy on RMFD.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MITREtrieval: Retrieving MITRE Techniques From Unstructured Threat Reports by Fusion of Deep Learning and Ontology

引用

IEEE Transactions on Network and Service Management 2024年第4期21卷 4871-4887页

作者： Yi-Ting Huang R. Vaitheeshwari Meng-Chang Chen Ying-Dar Lin Ren-Hung Hwang Po-Ching Lin Yuan-Cheng Lai Eric Hsiao-Kuang Wu Chung-Hsuan Chen Zi-Jie Liao Chung-Kuan Chen Department of Electrical Engineering National Taiwan University of Science and Technology Taipei Taiwan Department of Computer Science and Information Engineering National Central University Taoyuan Taiwan Institute of Information Science Academia Sinica Taipei Taiwan Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan College of Artificial Intelligence National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Computer Science and Information Engineering National Chung Cheng University Chiayi Taiwan Department of Information Management National Taiwan University of Science and Technology Taipei Taiwan Research Division CyCraft Technology New Taipei City Taiwan

Cyber Threat Intelligence (CTI) plays a crucial role in understanding and preemptively defending against emerging threats. Typically disseminated through unstructured reports, CTI encompasses detailed insights into threat actors, their actions, and attack patterns. The MITRE ATT&CK framework offers a comprehensive catalog of adversary tactics, techniques, and procedures (TTPs), serving as a valuable resource for deciphering attacker behavior and enhancing defensive measures. Addressing the challenge of time-consuming manual analysis of MITRE TTPs in unstructured CTI reports, this paper presents MITREtrieval, a novel system that leverages deep learning and ontology to efficiently extract MITRE techniques. This approach mitigates issues related to the implicit nature of TTPs, textual semantic dependencies, and the scarcity of adequately labeled datasets, enabling more effective analysis even with limited sample sizes. Our approach combines a sophisticated sentence-level BERT deep learning model with ontology knowledge to address sparse data challenges, using a voting algorithm to merge outcomes. This results in a more accurate classification of MITRE techniques, capturing contextual nuances effectively. Our evaluation confirms MITREtrieval’s effectiveness in identifying techniques, regardless of their representation in training samples. MITREtrieval has surpassed benchmarks, achieving F2 scores of 58%, 62%, and 69% in multi-label technique identification across 113, 46, and 23 CTI reports, respectively, thereby streamlining CTI analysis and improving threat intelligence.

关键词： computer security Cyber threat intelligence Ontologies Malware Natural language processing Deep learning Electronic mail

来源：评论

学校读者我要写书评

暂无评论

Advanced Multi-Spectral Image Processing Techniques for Enhanced Remote Sensing and Comprehensive Environmental Monitoring in Diverse Ecosystems

Advanced Multi-Spectral Image Processing Techniques for Enha...

引用

Recent Advances in science and engineering technology (ICRASET), International Conference on

作者： J. Jagannathan A. Ponmalar Rajkumar Pandiarajan I. Sudha P.S. Ramesh Department of Software and Systems Engineering School of Computer Science Engineering and Information Systems Vellore Institute of Technology Katpadi Vellore India R.M.K. Engineering College Chennai Tamilnadu India Department of Information and Technology School of Computing Kalasalingam Academy Of Research and Education Krishnankoil Virudhunagar India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences SIMATS University Chennai Tamil Nadu India Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai Tamilnadu India

ISBN: (数字)9798350388602

ISBN: (纸本)9798350388619

The research is aimed at improvement of multi-spectral image analysis for remote sensing and monitoring the environment at various ecosystems. A combined spectral imagery encompasses several spectrums at a time compared to single spectral imagery that takes only one spectrum at an instance or at a given area to cover all these spectrums. The study employs digital high resolution satellite images and ground based data to quantify the vegetation vigour, water quality, soil characteristics and changes in land cover. Features, indexes, and classification routines are employed to process the data that is collected with the aid of avionics and other peripherals. This provincial level modelling highlights important gains in environmental status, which includes the general improvement in vegetation condition, water quality as well as physical properties of the soil in the province. This study focuses on the complementary possibilities of integration of multi-spectral imaging with other remote sensing technologies and the high performance of data processing. These techniques play significant roles if used to give current and comprehensive information in the management and conservation of the environment as highlighted in the study.

关键词： Accuracy Wildfires Multispectral imaging Disasters Ecosystems Vegetation mapping Water quality Soil Sensors Remote sensing

来源：评论

学校读者我要写书评

暂无评论

A One Stage Non-motorized Vehicle Detection Network

A One Stage Non-motorized Vehicle Detection Network

引用

2022 International Conference on Cyber-Physical Social Intelligence, ICCSI 2022

作者： Fu, Xiaofeng Yuan, Xia Ye, Jialin Bo, Yuming School of Automation Nanjing University of Sicence and Technology Nanjing China Chengdu Southwest Information Control Research Institute Co. Ltd Chendu China School of Computer Science and Engineering Nanjing University of Sicence and Technology Nanjing China

ISBN: (数字)9781665498357

ISBN: (纸本)9781665498357

With the growth of non-motorized vehicles, low-carbon and environmentally friendly travel has gradually become the mainstream of people all over the world. The wide distribution and small proportion of non-motorized vehicles in the image make it difficult for traffic regulators to detection them, which increases the work of urban managers. Focusing on the task of non-motorized vehicle detection, we improve the YOLOv5s. The symmetric revolution module is used to construct the spatial attention mechanism, so as to improve the robustness of the network model. We design a spatial attention module to reduce the parameters. The proposed dimension decoupling module decouples the dimensions of the width and height of the feature map respectively, so that the network can adapt to the width and height of non-motorized vehicle targets. Through multi-scale receptive field fusion module, residual connection is used to improve the detection ability of the network to small targets. Finally, above modules are integrated into the YOLOv5s. Experiments on real road traffic images show that the proposed model has achieved good results in non-motorized vehicle detection. © 2022 IEEE.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

Development Progress of SWLBM a Framework Based on Lattice Boltzmann Method for Fluid Dynamics Simulation 1

引用

18th International Symposium on Applied Reconfigurable Computing, ARC 2022

作者： Xuesen, Chu Xiang, He Fang, Li Zhao, Liu Guangwen, Yang Department of Computer Science and Technology Tsinghua University Beijing100084 China China Ship Scientific Research Center Wuxi214082 China Taihu Laboratory of DeepSea Technological Science Wuxi214082 China Jiangnan Institute of Computing Technology Wuxi214083 China National Research Center of Parallel Computer Engineering and Technology Beijing100084 China National Supercomputing Center in Wuxi Wuxi214072 China

ISBN: (数字)9783031199837

ISBN: (纸本)9783031199820

SWLBM is a software framework based on Lattice Boltzmann Method (LBM) for Computational Fluid Dynamic (CFD) running on Sunway many-core processors. In this paper, we review the achievements of code developing in early stage and introduce the development progress recently of this software, including the development of parallel optimization for Sunway new-generation supercomputing system, functional extensions of the software like pre-process function for mesh generation from a geometry file with STL file and BMP file, Immersed boundary condition for moving subject simulation. Some applications with SWLBM will be introduced to show the advantage of this software over other CFD code in large-scale simulations. SWLBM is still under development, with the continuous improvement of functions, it will play a greater role in the field of fluid simulation. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

A study on the prediction model of dam seepage volume based on isolated forest-multiple stepwise linear regression 4

A study on the prediction model of dam seepage volume based ...

引用

4th International Conference on computer engineering and Application, ICCEA 2023

作者： Chen, Guanyu Zhang, Suli Zhang, Chenlin Qian, XiaoChun Changchun University of Technology School of Computer Science & Engineering Jilin Provincial Key Laboratory of Changbai Historical Culture and Vr Reconstruction Technology Jilin Prov. Changbai Mt. Ecol. Culture and Virtual Reality Simulation Engineering Research Center Changchun China Jilin Institute of Chemical Technology School of Information and Control Engineering Jilin Provincial Key Laboratory of Changbai Historical Culture and Vr Reconstruction Technology Jilin Prov. Changbai Mt. Ecol. Culture and Virtual Reality Simulation Engineering Research Center Changchun China

ISBN: (纸本)9798350347548

In the study of hydraulic seepage prediction, it is often necessary to target multiple feature dimensions. In this paper, a variety of traditional machine learning algorithms based on stepwise regression filtering features are used to train a dam seepage prediction model based on isolated forest processing of data outliers. The results show that the algorithm can be applied to predict the seepage data of a reservoir more accurately, and further suggest improvements to the dam's seepage control function and observation facilities. The experimental results show that the correlation coefficient of the linear regression model is significantly improved when the training sample ratio is 0.8 after the isolated forest treatment of outliers and stepwise regression screening features. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Brain Tumour Segmentation and Volume Estimation using Efficient Convolution Neural Network for MRI Images

Brain Tumour Segmentation and Volume Estimation using Effici...

引用

Distributed Computing and Electrical Circuits and Electronics (ICDCECE), IEEE International Conference on

作者： R Jaichandran R Shobana S Senthilkumar A Srinivasan Renuka Arora Ashok Kumar Department of Computer Science and Engineering Aarupadai Veedu Institute of Technology Organization Vinayaka Mission’s Research Foundation (Deemed to be University) Paiyanoor India Department of Computer Science and Engineering Jagannath University NCR Haryana Department of Computer Science and Engineering Banasthali Vidyapith Rajasthan India

A brain tumor is a type of blood clot in the form of cerebral disease. A brain tumor is a way to view and Magnetic Resonance Imaging (MRI) image detail is required. It is difficult to distinguish between normal tissue and tumor tissue from the brain. To should be an accurate analysis of brain tumours. Brain tumour segmentation is one of the most important and difficult tasks in the medical imaging field. Manual classification with human assistance can lead to incorrect predictions and diagnoses. Moreover, it is a difficult task when there is a large amount of data to help. Brain tumours are highly heterogeneous in appearance, and similarities between tumor and normal tissue make it difficult to extract tumour regions from images. In this paper, we proposed a method for extracting brain tumours from 2D magnetic resonance brain images (MRI) by a Fuzzy C-Means clustering algorithm with a traditional classifier and a convolutional neural network. Initially the pre-processing was carried out to apply the filter for image normalization with wiener filter and Robust active shape model was applied to predict the voluminous of data presence and segmentation was carried out through fuzzy c-means clustering and selected features was trained into optimize fuzzy c-means convolution neural network (FCM-CNN). This proposed system produces high detection accuracy in precision rate as well compared to other system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Security with Machine Learning-based Finger-Vein Biometric Authentication System

Enhancing Security with Machine Learning-based Finger-Vein B...

引用

Mobile Computing and Sustainable Informatics (ICMCSI), International Conference on

作者： G Prakash Babu Upendra Singh Aswal M. Sindhu Vijay Jagdish Upadhye Natrayan L Harshal Patil Department of Computer Science and Engineering Acharya Institute of Technology Bangalore Karnataka India Department of Computer Science & Engineering Graphic Era Deemed to be University Dehradun Uttarakhand India Department of Computer Science and Business Systems R.M.K. Engineering College Thiruvallur Tamil Nadu India Department of Microbiology Research and Development Cell Parul Institute of Applied Sciences Parul University Waghodia Gujarat India Department of Mechanical Engineering Saveetha School of Engineering SIMATS Chennai Tamil Nadu India Computer Science and Engineering Symbiosis Institute of Technology Symbiosis International (Deemed University) Pune Maharashtra India

Biometric authentication systems have gained popularity as the necessity for secure individual identification has expanded. Iris, hand geometry, fingerprints, retina, vein patterns on the fingers and palms, and voice patterns are all common biometric features. Biometric recognition systems based on finger-veins are becoming more common due to their numerous advantages, including low danger of duplication or fraud, minimum invasiveness, and excellent identification accuracy. Motivated by these advantages, our research looks into the subject of finger-vein authentication. Similar to identification, exploration of hidden patterns within data is essential for effective finger-vein recognition. The study employs two feature extraction approaches, the Local Binary Pattern (LBP) and the Gabor filter, as well as two machine learning models, the Adaptive Neuro-Fuzzy Inference System (ANFIS) and the Support Vector Machine (SVM). Using finger-vein images from the SDUMLA HMT database, this study evaluates four distinct feature extraction and ML model combinations. The simulation results highlight the excellent accuracy of 97.72%, False Acceptance Rate (FAR) of 3.1%, and False Rejection Rate (FRR) of 1.41% attained by LBP with ANFIS. In addition to performance measurements, the study examines the amount of time required for each task.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Estimating Effective Material Thermal Properties for Use in CFAST

Estimating Effective Material Thermal Properties for Use in ...

引用

2021 Symposium on Obtaining Data for Fire Growth Models

作者： Janssens, Marc L. Huczek, Jason P. Schluneker, Alexandra M. Icove, David J. Geosciences and Engineering Dept. Southwest Research Institute 6220 Culebra Rd. San AntonioTX78238 United States Fire Technology Dept. Southwest Research Institute 6220 Culebra Rd. San AntonioTX78238 United States Dept. of Electrical Engineering and Computer Science University of Tennessee 1520 Middle Dr. KnoxvilleTN37996 United States

ISBN: (纸本)9780803177314

A material thermal property database was developed for use in Consolidated Model of Fire and Smoke Transport (CFAST) simulations performed in support of fire investigations. The database includes thermal properties for interior surface materials;ignition, metal, and plastic target materials;and damage target materials. CFAST assumes that thermal properties of interior surface and target materials remain constant throughout the course of the fire. In reality, these properties vary with temperature. Assuming ambient temperature properties remain constant may result in significant errors in the heat conduction calculations. This paper presents the results of Fire Dynamics Simulator (FDS) compartment fire simulations to evaluate the effect on room temperature predictions of using constant versus temperature-dependent thermal properties for the gypsum board on walls and ceiling. The simulations show that assuming ambient temperature properties will overestimate the room temperature during the peak burning and decay periods of the fire. Fidelity of the predictions can be improved by increasing the specific heat in the constant property simulations to account for the gypsum dehydration energetics indirectly. The paper describes methodologies developed to estimate effective constant thermal properties for ignition and metal target materials and provides examples to illustrate their application. The methodology for ignition target materials relies on FDS heat conduction simulations of a series of ignition experiments conducted in the Cone Calorimeter. An evolutionary method is used to obtain values for thermal conductivity, specific heat, and ignition temperature that optimize agreement between calculated and measured ignition times. The approach to estimate effective thermal properties of metal target materials is similar. First, FDS and temperature-dependent property values of the metal are used to determine the time for the target surface to reach the melting point for a

关键词： Fires

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：