检索结果-内蒙古大学图书馆

Asia-Pacific Network Operations and Management Symposium (APNOMS)

作者： Hongyang Sun Huiling Shi Lei Zhang Lei Huang Wei Zhang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

With the increasing trend of network traffic encryption, both normal and abnormal traffic have begun to utilize encryption techniques to conceal their behavior, making accurate identification of encrypted traffic an essential prerequisite for network management and security. Current traffic classification methods generally suffer from inadequate feature extraction and lower multi-classification accuracy. This paper proposes a multiclassification hybrid deep learning model for encrypted traffic, called channel attention-based spatiotemporal convolutional neural network (CA-STCNN). The model extracts spatial features by convolutional neural network (CNN) and then introduces temporal convolutional neural network (TCN) to extract the temporal features. Using the convolution structure with more stable gradients can ensure the robustness of spatiotemporal feature extraction. Furthermore, the model also introduces the channel attention mechanism to assign weights to features in the channel dimension according to their contributions to classification. In the multi-classification experiments conducted using the ISCXVPN dataset, CA-STCNN outperforms existing methods in terms of accuracy, precision, recall, and F1 score.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimization Strategy for Remote Differential Upgrade of IoT Firmware

Optimization Strategy for Remote Differential Upgrade of IoT...

引用

IEEE Joint International information Technology and Artificial Intelligence Conference (ITAIC)

作者： Wenkang Ding Ruirui Sun Qinglong Meng Zhong Wan Guanghe Cheng Huijuan Hao Jinqiang Bai Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

With the continuous increase of IoT devices, the management and optimization of devices become particularly important. The update cycle of device functions is short, and in order to facilitate management and optimize device performance, a fast and convenient upgrade method is needed to solve this problem. Remote upgrade technology is mainly divided into full package upgrade and differential upgrade. Compared to full package upgrade, differential upgrade has obvious advantages in remote upgrade of large files. The main problem with current differential upgrade is the problem of slow generation speed while ensuring a certain size range of differential packages. This article proposes an improved Bsdiff algorithm to solve the problem of slow generation speed of differential packages. Experimental verification shows that the efficiency of remote upgrade has been improved, Improved the speed of differential packet generation and upgrade efficiency while ensuring the size of differential packets.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Malicious Traffic Classification for IoT based on Graph Attention Network and Long Short-Term Memory Network

Malicious Traffic Classification for IoT based on Graph Atte...

引用

Asia-Pacific Network Operations and Management Symposium (APNOMS)

作者： Lei Zhang Lizhuang Tan Huiling Shi Hongyang Sun Wei Zhang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan China

With the rapid development of digital services and the Internet of Things (IoT), the number and diversity of cyber attacks have reached unprecedented levels, making network traffic classification a crucial means of maintaining network security. However, in the face of increasingly complex traffic attacks, traditional analysis of individual traffic flows is no longer sufficient for effective detection. Therefore, to reveal the potential correlations between network traffic flows, we propose a novel IoT malicious traffic classification model called EGAT-LSTM. This model combines an improved Graph Attention Network (GAT) with Long Short-Term Memory (LSTM) network to effectively capture the spatial topology and temporal features of network traffic, ultimately achieving efficient classification. Moreover, we optimize the aggregation function of network node features within the attention mechanism to effectively address the issue of essential information retention during the feature learning process. The experimental results on two publicly available IoT datasets demonstrate that our method achieves improvements in both binary and multi-class classification tasks, exhibiting significant advantages over some current approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Language-Guided Face Animation by Recurrent StyleGAN-based Generator

arXiv

引用

arXiv 2022年

作者： Hang, Tiankai Yang, Huan Liu, Bei Fu, Jianlong Geng, Xin Guo, Baining The School of Computer Science and Engineering The Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China Microsoft Research Asia China

Recent works on language-guided image manipulation have shown great power of language in providing rich semantics, especially for face images. However, the other natural information, motions, in language is less explored. In this paper, we leverage the motion information and study a novel task, language-guided face animation, that aims to animate a static face image with the help of languages. To better utilize both semantics and motions from languages, we propose a simple yet effective framework. Specifically, we propose a recurrent motion generator to extract a series of semantic and motion information from the language and feed it along with visual information to a pre-trained StyleGAN to generate high-quality frames. To optimize the proposed framework, three carefully designed loss functions are proposed including a regularization loss to keep the face identity, a path length regularization loss to ensure motion smoothness, and a contrastive loss to enable video synthesis with various language guidance in one single model. Extensive experiments with both qualitative and quantitative evaluations on diverse domains (e.g., human face, anime face, and dog face) demonstrate the superiority of our model in generating high-quality and realistic videos from one still image with the guidance of language. Code will be available at https://***/TiankaiHang/*** Copyright © 2022, The Authors. All rights reserved.

关键词： Animation

来源：评论

学校读者我要写书评

暂无评论

enCBS: Delay Guarantee Mechanism based on Credit-based Shaping for Programmable Switch

enCBS: Delay Guarantee Mechanism based on Credit-based Shapi...

引用

Asia-Pacific Network Operations and Management Symposium (APNOMS)

作者： Xinhang Wang Lizhuang Tan Wei Zhang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

The Time-Sensitive Networking (TSN) aims to achieve delay guarantee through traffic shaping, and the programmable switch provides a data plane implementation solution for the flexible and deployable TSN network. However, the built-in quality of service guarantee mechanism of the current programmable switch cannot satisfy the deterministic guarantee of queue control and end-to-end delay of TSN network. Credit-based Shaping (CBS) is a common traffic shaping scheme for static TSN network, and it is helpless in the face of topology changes and dynamic entry/exit of flows. In this paper, we design the enhance CBS (enCBS), a CBS-based delay guarantee mechanism on the programmable switch, using the advantages of Software-Defined Network (SDN) architecture to optimize flow management and parameter configuration. In addition, we improved the traditional CBS and strengthened the packet interval constraints. The experimental results prove that, compared with the existing quality of service control mechanism of programmable switch, the delay guarantee mechanism based on CBS shaping has better delay guarantee performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LoMAE: Low-level Vision Masked Autoencoders for Low-dose CT Denoising

arXiv

引用

arXiv 2023年

作者： Wang, Dayang Xu, Yongshun Han, Shuo Wu, Zhan Zhou, Li Morovati, Bahareh Yu, Hengyong The Department of Electrical and Computer Engineering University of Massachusetts Lowell LowellMA01854 United States The Laboratory of Image Science and Technology Southeast University Nanjing210096 China The Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing210096 China

Low-dose computed tomography (LDCT) offers reduced X-ray radiation exposure but at the cost of compromised image quality, characterized by increased noise and artifacts. Recently, transformer models emerged as a promising avenue to enhance LDCT image quality. However, the success of such models relies on a large amount of paired noisy and clean images, which are often scarce in clinical settings. In the fields of computer vision and natural language processing, masked autoencoders (MAE) have been recognized as an effective label-free self-pretraining method for transformers, due to their exceptional feature representation ability. However, the original pretraining and fine-tuning design fails to work in low-level vision tasks like denoising. In response to this challenge, we redesign the classical encoder-decoder learning model and facilitate a simple yet effective low-level vision MAE, referred to as LoMAE, tailored to address the LDCT denoising problem. Moreover, we introduce an MAE-GradCAM method to shed light on the latent learning mechanisms of the MAE/LoMAE. Additionally, we explore the LoMAE’s robustness and generability across a variety of noise levels. Experiments results show that the proposed LoMAE can enhance the transformer’s denoising performance and greatly relieve the dependence on the ground truth clean data. It also demonstrates remarkable robustness and generalizability over a spectrum of noise levels. © 2023, CC BY.

关键词： Image quality

来源：评论

学校读者我要写书评

暂无评论

A Global Task Scheduling Method Based on Network Measurement and Prediction in Computing Power networks

A Global Task Scheduling Method Based on Network Measurement...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Chuanfu Zhang Wen Li Jing Chen Jing Ge Di Wang Na Li Hao Sun Yudong Geng Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

To improve the task scheduling efficiency in computing power networks, this paper proposes a global task scheduling method based on network measurement and prediction in computing power networks (GTS-MP), which selects a more suitable computing power cluster and storage platform with a higher network availability score between them to implement tasks. The experiments are conducted on a test bench of urban area networks in 12 cities. The results show that this method greatly reduces the execution time of artificial intelligence model training tasks, with a maximum reduction of 40% compared to other plans.

关键词：

来源：评论

学校读者我要写书评

暂无评论

L2fusion: Low-Light Oriented Infrared and Visible Image Fusion

L2fusion: Low-Light Oriented Infrared and Visible Image Fusi...

引用

IEEE International Conference on Image Processing

作者： Xiang Gao Guohua Lv Aimei Dong Zhonghe Wei Jinyong Cheng Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

Infrared and visible image fusion aims to integrate salient targets and abundant texture information into a single fused image. Existing methods typically ignore the issue of illumination, so that there are problems of weak texture details and poor visual perception in case of low illumination. To address this issue, we propose a low-light oriented infrared and visible image fusion network, named L2Fusion. In particular, we first design a decomposition network according to Retinex theory to obtain the reflectance features of a visible image with low-light. Then, these features are integrated with the features extracted from the corresponding infrared image by a residual network. The finally fused image largely eliminates the negative impact caused by low illumination, and contains both salient targets and abundant texture information. Extensive experiments demonstrate the superiority of our L2Fusion over the state-of-the-art methods, in terms of both visual effect and quantitative metrics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Prototype-based classifier learning for long-tailed visual recognition

引用

Science China(information Sciences) 2022年第6期65卷 62-76页

作者： Xiu-Shen WEI Shu-Lin XU Hao CHEN Liang XIAO Yuxin PENG School of Computer Science and Engineering Nanjing University of Science and Technology State Key Laboratory of Integrated Services Networks Xidian University Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education and Jiangsu Key Lab of Image and Video Understanding for Social Security Wangxuan Institute of Computer Technology Peking University

In this paper, we tackle the long-tailed visual recognition problem from the categorical prototype perspective by proposing a prototype-based classifier learning(PCL) method. Specifically, thanks to the generalization ability and robustness, categorical prototypes reveal their advantages of representing the category semantics. Coupled with their class-balance characteristic, categorical prototypes also show potential for handling data imbalance. In our PCL, we propose to generate the categorical classifiers based on the prototypes by performing a learnable mapping function. To further alleviate the impact of imbalance on classifier generation, two kinds of classifier calibration approaches are designed from both prototype-level and example-level aspects. Extensive experiments on five benchmark datasets, including the large-scale iNaturalist, Places-LT,and ImageNet-LT, justify that the proposed PCL can outperform state-of-the-arts. Furthermore, validation experiments can demonstrate the effectiveness of tailored designs in PCL for long-tailed problems.

关键词： long-tailed distribution categorical prototype classifier generation classifier calibration class imbalance

来源：评论

学校读者我要写书评

暂无评论

Felix: A Model of Detecting Off-chain Abnormal States in Decentralized Applications

Felix: A Model of Detecting Off-chain Abnormal States in Dec...

引用

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Lianhai Wang Qihao Huang Wei Shao Jinpeng Wang Xiaoqian Liu Fansheng Wang Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Key Laboratory of Computing Power Network and Information Security Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350381993

ISBN: (纸本)9798350382006

With the rapid development of blockchain technology, the use of decentralized applications (DApps) has experienced significant growth. However, current testing methods for DApps primarily focus on testing the smart contracts on the blockchain, which are the foundation of DApps, but lack a comprehensive approach to effectively detect off-chain abnormal states. To address this issue, this paper proposes a generic abnormal state detection model based on off-chain transaction data. The model leverages the DApp program code logic to set up test oracles to analyze off-chain transaction data. Experimental results demonstrate that the model achieves high accuracy in detecting off-chain abnormal states, with a prediction accuracy as high as 96.3%. Furthermore, a comparison with other related vulnerability detection methods shows the advantages of the proposed approach.

关键词： Analytical models Technological innovation Privacy Smart contracts Decentralized applications Data models Blockchains

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：