检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zhou, Ziqi Hu, Shengshan Zhao, Ruizhi Wang, Qian Zhang, Leo Yu Hou, Junhui Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Cyber Science and Engineering Wuhan University China School of Information and Communication Technology Griffith University Australia Department of Computer Science City University of Hong Kong Hong Kong National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial use. In this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://github. com/CGCL-codes/AdvEncoder. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A Security Data Sharing Model for Cloud Network 6th

A Security Data Sharing Model for Cloud Network

引用

6th International Conference on Data Mining and Big Data, DMBD 2021

作者： Zhao, Kunyuan Zhang, Qikun Wu, Yimeng Zhu, Liang Zhu, Hongfei Wang, Ruifang School of Computer and Communication Engineering Zhengzhou University of Light Industry Zhengzhou450001 China Zhengzhou Technical College Zhengzhou450000 China College of Computer and Technology Chongqing University of Posts and Telecommunications Chongqing400065 China Department of System Technology Zhengzhou University of Light Industry Zhengzhou450000 China

ISBN: (纸本)9789811675010

With the continuous development and wide application of cloud computing technology, its security threats are also increasing, especially in data security sharing and user privacy security protection. Therefore, it is very necessary to establish a security and reliable data sharing model. Combined with blockchain technology, this paper proposes a security data sharing model for cloud network. Blockchain has the characteristics of non-forgery and traceability. By storing the user identity information into the blockchain, not only the authentication efficiency of the user identity can be improved, but also the security of the data sharing system can be improved. At the same time, bilinear mapping and attribute signature technology are used to realize mutual authentication among different entities in domains, secure data sharing and secure interoperability among domains are protected. In addition, the sharing model has the characteristics of low network delay, supporting massive data access and elastic infrastructure. After proof and analysis, the scheme has better security and performance than other references. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

A Novel Ensemble Deep Learning Framework for Breast Cancer Prediction

A Novel Ensemble Deep Learning Framework for Breast Cancer P...

引用

Innovative Computing, Intelligent communication and Smart Electrical systems (ICSES), International Conference on

作者： Pundru Chandra Shaker Reddy S Nithyapriya Nipun Sharma S Maheswari B Jayaram Tejaswini Katale School of Computer Science and Artificial Intelligence SR University Warangal Telangana India Artificial Intelligence and Data Science Bannari Amman Institute of Technology Anna University Chennai Tamil Nadu India Electronics and Communication Engineering Presidency University Bangalore Karnataka India Department of Mathematics Sathyabama Institute of Science and Technology Chennai Tamil Nadu India Computer Science and Business System RMK Engineering College Kavaraipettai Tamil Nadu India Department of Computer Engineering Vishwakarma Institute of Technology Pune Maharashtra India

Because of its high incidence and high mortality rate, breast cancer requires reliable means of early detection and diagnosis to improve survival rates for its patients. With advantages including increased accuracy and efficiency, deep learning (DL) and machine learning (ML) methods are quickly becoming indispensable in the fight against breast cancer. However, there is a need for additional study because existing approaches have scalability and performance constraints. Utilizing a pre-trained ResNet50V2 and ensemble-based ML approaches, we offer a hybrid approach to reliable breast cancer diagnosis in this study. Learning and extracting latent trends from complex breast-cancer(BC) images is made possible by DL's incorporation, with ML strategies contributing interpretability and generalizability. We performed wide experiments utilizing a openly accessible Invasive-Ductal-Carcinoma(IDC) dataset based on breast histopathology images of varying sample sizes. The resilience and great performance of our hybrid model are supported by the results of our extensive experiments. When compared to state-of-the-art models, our 95% accuracy rate, 94.5 precision, 94.2 recall, and 94.6 F1 score are all improvements. We also found that the best ML model to use with the ResNet50V2 architecture is the Light Boosting Classifier (LGB). This study's findings provide substantial contributions to the field of breast cancer diagnosis thanks to its novel approach, in-depth performance analysis, and trustworthy evaluation. In addition, it may help doctors make better decisions, provide better care, and boost results for people with breast cancer.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hardware design and tests of SMURF V1 platform for searching survivors in debris cones 18

Hardware design and tests of SMURF V1 platform for searching...

引用

18th International Conference on Information systems for Crisis Response and Management, ISCRAM 2021

作者： Watanabe, Masahiro Takahashi, Kenichi Kimura, Tetsuya Tadakuma, Kenjiro Mugaas, Terje Ozawa, Yu Takane, Eri Suzuki, Soichiro Marafioti, Giancarlo Koutsokeras, Miltiadis Tadokoro, Satoshi Tough Cyberphysical AI Research Center Tohoku University Japan Department of System Safety Nagaoka University of Technology Japan Department of Mathematics and Cybernetics SINTEF Digital Norway Graduate School of Information Sciences Tohoku University Japan Graduate School of Engineering Tohoku University Japan International Rescue System Institute United States Institute of Communication and Computer Systems Greece

ISBN: (纸本)9781949373615

When a large-scale disaster such as earthquake occurs, a huge number of victims will be trapped under debris in a wide area. Rescue activities in debris are technically not easy and endanger the first responders. There are several methods for improving safety and efficiency of rescue operation, but their availability is limited to a certain area or short operating time. Our project called CURSOR is developing tools to comprehensively search victims using a large number of ground-based robots entering debris transported by aerial drones. Here we show the development of the exploration robot collecting information with several sensors. The robot system was designed based on the requirements and performance was evaluated by ruggedization tests and mobility tests. No critical problem was found in the durability, and the mobility showed as the same as the ordinary wheel. To improve the mobility, we are planning to apply a proposed unique track mechanism. © 2021 Information systems for Crisis Response and Management, ISCRAM. All rights reserved.

关键词： Machine design

来源：评论

学校读者我要写书评

暂无评论

Multi-modal Data based Semi-Supervised Learning for Vehicle Positioning

arXiv

引用

arXiv 2024年

作者： Huan, Ouwen Yang, Yang Luo, Tao Chen, Mingzhe Beijing Laboratory of Advanced Information Network Beijing University of Posts and Telecommunications Beijing100876 China Department of Electrical and Computer Engineering Institute for Data Science and Computing University of Miami Coral GablesFL33146 United States Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China

In this paper, a multi-modal data based semisupervised learning (SSL) framework that jointly use channel state information (CSI) data and RGB images for vehicle positioning is designed. In particular, an outdoor positioning system where the vehicle locations are determined by a base station (BS) is considered. The BS equipped with several cameras can collect a large amount of unlabeled CSI data and a small number of labeled CSI data of vehicles, and the images taken by cameras. Although the collected images contain partial information of vehicles (i.e. azimuth angles of vehicles), the relationship between the unlabeled CSI data and its azimuth angle, and the distances between the BS and the vehicles captured by images are both unknown. Therefore, the images cannot be directly used as the labels of unlabeled CSI data to train a positioning model. To exploit unlabeled CSI data and images, a SSL framework that consists of a pretraining stage and a downstream training stage is proposed. In the pretraining stage, the azimuth angles obtained from the images are considered as the labels of unlabeled CSI data to pretrain the positioning model. In the downstream training stage, a small sized labeled dataset in which the accurate vehicle positions are considered as labels is used to retrain the model. Simulation results show that the proposed method can reduce the positioning error by up to 30% compared to a baseline where the model is not pretrained. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Research on Restructuring Path of Children's Digital Publishing Supply and Demand Chain 2

Research on Restructuring Path of Children's Digital Publish...

引用

2nd International Conference on Information Science and Education, ICISE-IE 2021

作者： Chen, Huan Liu, Ying Shan, Chan Zhang, Hequn School of Journalism and Communication of Minjiang University Fujian Fuzhou350108 China Department of Software Technology Xiamen Institute of Software Technology Fujian Xiamen China Department of Computer System and Technology Faculty of Computer Science and Information Technology Universiti Malaya Kuala Lumpur50603 Malaysia College of Information Engineering Jimei University Xiamen361021 China

ISBN: (纸本)9781665438292

In the era of rapid development of intelligent technology, the children's digital publishing industry is facing a disruptive change, and this change may be realized through the reconstruction of the supply and demand chain. The value power of source content creation and the ability to gather users have been deconstructed. The increase in the amount of content requires platform institutions to have a more professional content operation system. The convenience of payment technology has greatly increased the digital content payment rate of users of children's digital products. A development model of the supply and demand chain of children's digital publishing with more innovative flexibility and enabling effects. © 2021 IEEE

关键词： Economics

来源：评论

学校读者我要写书评

暂无评论

Hybrid MPIE Method with Anisotropic Impedance Boundary Conditions for Layered Media

Hybrid MPIE Method with Anisotropic Impedance Boundary Condi...

引用

International Applied Computational Electromagnetics Society Symposium (ACES)

作者： Ying Zhao Wenjie Neng Zhen Zhang Dazhi Ding College of computer and information engineering (College of artificial intelligence) Nanjing Tech University Nanjing China State Key Laboratory of Millimeter Waves Nanjing China China Key System & Integrated Circuit Co. Ltd. Wuxi China Department of communication engineering Nanjing University of Science and Technology Nanjing China

A mixed potential integral equation (MPIE) method combined with anisotropic impedance boundary conditions is proposed for the analysis of scattering from layered media. In this method, anisotropic surface conductivity is modeled by the impedance boundary conditions. As a result, only these surfaces need to be discretized. Numerical example is demonstrated to show the accuracy of this method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Correction to: A new approach to snow avalanche rescue using UAV pictures based on convolutional neural networks

引用

Journal of Real-Time Image Processing 2023年第5期20卷 1-2页

作者： Zhang, Suyu Gavrilovskaya, Nadezhda Al Said, Nidal Afandi, Waleed Saeed College of Artificial Intelligence Wenzhou Polytechnic Wenzhou China Department of Computer-Aided Design Systems and Engineering Calculations Moscow Timiryazev Agricultural Academy Russian State Agrarian University Moscow Russian Federation College of Mass Communication Ajman University Ajman United Arab Emirates Department of Information System King Abdulaziz University Jeddah Kingdom of Saudi Arabia

来源：评论

学校读者我要写书评

暂无评论

A 180-Degree Phase Difference Output of Wilkinson Power Divider Using Microstrip Parallel Coupled Lines

A 180-Degree Phase Difference Output of Wilkinson Power Divi...

引用

Conference on Power, Energy and Innovations (ICPEI)

作者： Pratya Phanomupatum Ravee Phromloungsri Rachata Tama Tiwat Pongthavornkamol Decha Wongpatsa Somchat Sonasang Department of Electronics Faculty Industrial Technology Nakhon Phanom University Nakhon Phanom Thailand Computer and Communication Engineering Faculty of Technology Udon Thani Rajabhat University Udon Thani Thailand Location and Automatic Identification System Research Team Communication and Network Research Group National Electronic and Computer Technology Center National Science and Technology Development Agency Pathumthani Thailand Electrical Elements and Wireless Devices Unit (EEWSU) Faculty Industrial Technology Nakhon Phanom University Nakhon Phanom Thailand

ISBN: (数字)9798350356779

ISBN: (纸本)9798350356786

The design of a 180° output phase difference Wilkinson power divider for L-band applications is detailed in this article. The Wilkinson power is generated through the utilization of the parallel coupler's phase shifter. To showcase the potential of the suggested power divider, an FR4 microwave substrate is fabricated with a narrow band of 1.6 GHz to accommodate radio astronomy systems. The insertion loss, return loss, and isolation loss for the power divider's difference phase are accurately simulated at the operational frequency.

关键词： Power dividers Technological innovation Microwave integrated circuits Radio astronomy Phase shifters Insertion loss Microwave circuits Microstrip Proposals Substrates

来源：评论

学校读者我要写书评

暂无评论

Enhancing Chewing Detection with a CNN-BiLSTM Network Utilizing IMU and Photoplethysmography Data from an Earable Device

Enhancing Chewing Detection with a CNN-BiLSTM Network Utiliz...

引用

IEEE International Conference on Software engineering and Service Sciences (ICSESS)

作者： Sakorn Mekruksavanich Wikanda Phaphan Phaisarn Jeefoo Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication Technology University of Phayao Phayao Thailand Research Group in Statistical Learning and Inference Department of Applied Statistics Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand Geographic Information System School of Information and Communication Technology University of Phayao Phayao Thailand Intelligent and Nonlinear Dynamic Innovations Research Center Department of Mathematics Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798350389166

ISBN: (纸本)9798350389173

Detecting chewing is vital to tracking eating habits, which is necessary to preserve good well-being and health. Advancements in wearable sensors have recently enabled cap-turing a wide range of sensor data, including signals from inertial measurement units (IMUs) and photoplethysmography (PPG). This research presents a new deep-learning design that efficiently detects chewing occurrences employing IMU and PPG sensor data acquired from an early instrument. The design includes convolutional neural networks (CNN) and bidirectional long short-term memory (LSTM). The CNN-BiLSTM network presented in this study combines the advantages of both CNN and BiLSTM to capture spatial and temporal characteristics from the sensor data effectively. The CNN layers extract localized patterns and characteristics from the IMU and PPG data, whereas the BiLSTM layers represent the extended relationships and time-based changes in chewing behavior. By integrating these two formidable designs, our network attains a high level of resilience and precision in detecting chewing. We assess the efficacy of the CNN-BiLSTM network using a standardized dataset obtained from several individuals wearing wearable devices while engaging in eating events. The dataset comprises IMU and PPG sensor data that have been annotated with associated chewing sessions. The experimental findings clearly show that the suggested network outperforms the current leading approaches, obtaining an accu-racy of 98.18 % and an F1-score of 98.29 %, the highest recorded. The CNN-BiLSTM network performs better than baseline models such as CNN, LSTM, BiLSTM, GRU, and BiGRU, emphasizing the efficacy of integrating spatial and temporal modeling for chewing identification.

关键词： Performance evaluation Recurrent neural networks Measurement units Photoplethysmography Convolutional neural networks Wearable devices Long short term memory Wearable sensors Software engineering Resilience

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：