检索结果-内蒙古大学图书馆

32nd ACM World Wide Web Conference, WWW 2023

作者： Dong, Zeqian He, Qiang Chen, Feifei Jin, Hai Gu, Tao Yang, Yun School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computing Technologies Swinburne University of Technology Australia School of Information Technology Deakin University Australia School of Computing Macquarie University Australia National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9781450394161

Training machine learning (ML) models on mobile and Web-of-Things (WoT) has been widely acknowledged and employed as a promising solution to privacy-preserving ML. However, these end-devices often suffer from constrained resources and fail to accommodate increasingly large ML models that crave great computation power. Offloading ML models partially to the cloud for training strikes a trade-off between privacy preservation and resource requirements. However, device-cloud training creates communication overheads that delay model training tremendously. This paper presents EdgeMove, the first device-edge training scheme that enables fast pipelined model training across edge devices and edge servers. It employs probing-based mechanisms to tackle the new challenges raised by device-edge training. Before training begins, it probes nearby edge servers' training performance and bootstraps model training by constructing a training pipeline with an approximate model partitioning. During the training process, EdgeMove accommodates user mobility and system dynamics by probing nearby edge servers' training performance adaptively and adapting the training pipeline proactively. Extensive experiments are conducted with two popular DNN models trained on four datasets for three ML tasks. The results demonstrate that EdgeMove achieves a 1.3 × -2.1 × speedup over the state-of-the-art scheme. © 2023 ACM.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Seat Occupancy Detection in Electric Vehicles: Unveiling Passenger Presence through Overlap Area Analysis

Seat Occupancy Detection in Electric Vehicles: Unveiling Pas...

引用

International Electronics Symposium (IES)

作者： Alivia Haris Alfan Rizaldy Pratama Bima Sena Bayu Dewantara Dadet Pramadihanto Center for Research and Innovation on Advanced Transportation Electrification Politeknik Elektronika Negeri Surabaya Surabaya Indonesia Data Science Department Faculty of Computer Science Universitas Pembangunan Nasional Veteran Jawa Timur Surabaya Indonesia Department of Informatics and Computer Engineering Politeknik Elektronika Negeri Surabaya Surabaya Indonesia

ISBN: (数字)9798350391992

ISBN: (纸本)9798350392005

The high number of bus passengers often leads to situations where some passengers are unable to secure a seat, resulting in the inconvenience of standing for the whole ride or wasting time waiting at the bus stop. To address this issue, this paper aims to develop a system capable of detecting bus seat occupancy, differentiating between occupied and unoccupied seats by passengers to inform potential passengers if seats are available before they take the next bus and help the government monitor its citizen and improve its public transportation services. The system uses a YOLO algorithm to perform object detection, enabling it to detect passengers, and an overlap percentage formula between passenger and seat bounding box to determine if the detected passenger is occupying the seat. The system provides real-time video output with texts indicating the total number of current passengers, total boarded passengers, also occupied and unoccupied seats. The system achieves accuracy of 95% for the front seat section and 92% for the back seat section.

关键词： YOLO Accuracy Government Streaming media Electric vehicles Real-time systems Distance measurement

来源：评论

学校读者我要写书评

暂无评论

PipeEdge: A Trusted Pipelining Collaborative Edge Training based on Blockchain 23

PipeEdge: A Trusted Pipelining Collaborative Edge Training b...

引用

32nd ACM World Wide Web Conference, WWW 2023

作者： Yuan, Liang He, Qiang Chen, Feifei Dou, Ruihan Jin, Hai Yang, Yun School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computing Technologies Swinburne University of Technology Australia School of Information Technology Deakin University Australia Faculty of Mathematics University of Waterloo Canada National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9781450394161

Powered by the massive data generated by the blossom of mobile and Web-of-Things (WoT) devices, Deep Neural Networks (DNNs) have developed both in accuracy and size in recent years. Conventional cloud-based DNN training incurs rapidly-increasing data and model transmission overheads as well as privacy issues. Mobile edge computing (MEC) provides a promising solution by facilitating DNN model training on edge servers at the network edge. However, edge servers often suffer from constrained resources and need to collaborate on DNN training. Unfortunately, managed by different telecoms, edge servers cannot properly collaborate with each other without incentives and trust. In this paper, we introduce PipeEdge, a scheme that promotes collaborative edge training between edge servers by introducing incentives and trust based on blockchain. Under the PipeEdge scheme, edge servers can hire trustworthy workers for pipelined DNN training tasks based on model parallelism. We implement PipeEdge and evaluate it comprehensively with four different DNN models. The results show that it outperforms state-of-the-art schemes by up to 173.98% with negligible overheads. © 2023 ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Using Image Processing for Determination of Some Physical Properties of Nano-materials

Using Image Processing for Determination of Some Physical Pr...

引用

International Conference on Signal Processing and Communication (ICSC)

作者： M. A. Ashabrawy Department of Computer Science College of Computer Engineering and Science in AI-Kharj Prince Sattam bin AbdulAziz University AI-Kharj Saudi Arabia Reactors Department Nuclear Research Center Atomic Energy Authority Egypt

ISBN: (纸本)9781665454315

In this article we will process the images obtained through nanoscale devices to obtain some physical properties of nanomaterials by using image processing techniques, numerical methods, before we do an improvement of the images obtained for a precise image high clarity and then we use numerical methods which is a method of completion through which you can get all the physical properties of samples and others, Images, which obtained through nanoscale devices as the transmission electron microscope (TEM) to obtain some physical properties of nanomaterials by using image processing techniques, numerical methods as Cubic Convolution Interpolation (CCI), We will improve the images obtained for accurate image High clarity and then we use numerical methods, which is a method of interpolation method through which you can get all the physical properties of Nano-materials, We will provide a general introduction by obtaining nanoscale images, numerical methods image-processing methods and many other important aspects of high resolution imaging, for example pictures of glycerin and other materials to illustrate a variety of as the transmission electron microscope (TEM) applications to study nanomaterials. Finally, We will obtain set of images by Nano device as the transmission electron microscope (TEM) or any other device, for samples prepared samples in diverse conditions having different physical properties, and analyzed this a set of images using some numerical technique in image processing, which starting by converting images to data file $(^{\ast}.\mathrm{dat})$ using programming languages codes. After using filters and we will get the more results some physical properties of Nano-materials.

关键词： Interpolation Codes Convolution Transmission electron microscopy Image processing Microscopy Signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

Adaptive fast-reaching nonsingular terminal sliding mode tracking control for quadrotor UAVs subject to model uncertainties and external disturbances

引用

ain Shams engineering Journal 2023年第8期14卷

作者： Mobayen, Saleh El-Sousy, Fayez F.M. Alattas, Khalid A Mofid, Omid Fekih, Afef Rojsiraphisal, Thaned Graduate School of Intelligent Data Science National Yunlin University of Science and Technology 123 University Road Section 3 Douliou Yunlin640301 Taiwan Department of Electrical Engineering Prince Sattam Bin Abdulaziz University Al Kharj Saudi Arabia Department of Computer Science and Artificial Intelligence College of Computer Science and Engineering University of Jeddah Jeddah Saudi Arabia Department of Electrical and Computer Engineering University of Louisiana at Lafayette Lafayette United States Advanced Research Center for Computational Simulation Department of Mathematics Faculty of Science Chiang Mai University Chiang Mai50200 Thailand Data Science Research Center Department of Mathematics Faculty of Science Chiang Mai University Chiang Mai50200 Thailand

This paper proposes an adaptive fast-reaching nonsingular terminal sliding mode control approach for the position and attitude tracking control of a quadrotor UAV subject to model uncertainty and external disturbances. To this end, a fast-reaching nonsingular terminal sliding surface is defined with the aim of fast convergence of the tracking error. The fast reachability of the nonsingular terminal sliding surface is proven using the Lyapunov stability theory. Since, in practice, the upper bounds of model uncertainties and external disturbances are unknown, an adaptive control scheme is considered to approximate them. The fast-reaching nonsingular terminal sliding mode control approach was augmented with the adaptive scheme to ensure both, the fast convergence of the switching surfaces and the rejection of model uncertainties and external disturbances. The effectiveness of the proposed adaptive fast-reaching nonsingular terminal sliding mode control approach is assessed using simulations of a quadrotor system implemented in MATLAB/Simulink software. The obtained results are further compared to those of an existing approach in terms of tracking performance, robustness, and integrator of absolute-value of error criteria. © 2022 THE AUTHORS

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Zero-Shot Skeleton-based Action Recognition with Dual Visual-Text Alignment

arXiv

引用

arXiv 2024年

作者： Kuang, Jidong Wang, Hongsong Han, Chaolei Gui, Jie The School of Cyber Science and Engineering Southeast University Nanjing210000 China Department of Computer Science and Engineering Southeast University Nanjing210096 China The Purple Mountain Laboratories Nanjing210000 China The Engineering Research Center of Blockchain Application Supervision and Management Ministry of Education Southeast University Nanjing210000 China

Zero-shot action recognition, which addresses the issue of scalability and generalization in action recognition and allows the models to adapt to new and unseen actions dynamically, is an important research topic in computer vision communities. The key to zero-shot action recognition lies in aligning visual features with semantic vectors representing action categories. Most existing methods either directly project visual features onto the semantic space of text category or learn a shared embedding space between the two modalities. However, a direct projection cannot accurately align the two modalities, and learning robust and discriminative embedding space between visual and text representations is often difficult. To address these issues, we introduce Dual Visual-Text Alignment (DVTA) for skeleton-based zero-shot action recognition. The DVTA consists of two alignment modules-Direct Alignment (DA) and Augmented Alignment (AA)-along with a designed Semantic Description Enhancement (SDE). The DA module maps the skeleton features to the semantic space through a specially designed visual projector, followed by the SDE, which is based on cross-attention to enhance the connection between skeleton and text, thereby reducing the gap between modalities. The AA module further strengthens the learning of the embedding space by utilizing deep metric learning to learn the similarity between skeleton and text. Our approach achieves state-of-the-art performances on several popular zero-shot skeleton-based action recognition benchmarks. Copyright © 2024, The Authors. All rights reserved.

关键词： Musculoskeletal system

来源：评论

学校读者我要写书评

暂无评论

Evaluating Pre-trained Convolutional Neural Networks and Foundation Models as Feature Extractors for Content-based Medical Image Retrieval

arXiv

引用

arXiv 2024年

作者： Mahbod, Amirreza Saeidi, Nematollah Hatamikia, Sepideh Woitek, Ramona Research Center for Medical Image Analysis and Artificial Intelligence Department of Medicine Faculty of Medicine and Dentistry Danube Private University Krems an der Donau Austria Artificial Intelligence Department Faculty of Computer Engineering University of Isfahan Isfahan Iran Research Center for Clinical AI-Research in Omics and Medical Data Science Department of Medicine Faculty of Medicine and Dentistry Danube Private University Krems an der Donau Austria Austrian Center for Medical Innovation and Technology Wiener Neustadt Austria

Medical image retrieval refers to the task of finding similar images for given query images in a database, with applications such as diagnosis support. While traditional medical image retrieval relied on clinical metadata, content-based medical image retrieval (CBMIR) depends on image features, which can be extracted automatically or semi-automatically. Many approaches have been proposed for CBMIR, and among them, using pre-trained convolutional neural networks (CNNs) is a widely utilized approach. However, considering the recent advances in the development of foundation models for various computer vision tasks, their application for CBMIR can also be investigated. In this study, we used several pre-trained feature extractors from well-known pre-trained CNNs and pre-trained foundation models and investigated the CBMIR performance on eight types of two-dimensional (2D) and three-dimensional (3D) medical images. Furthermore, we investigated the effect of image size on the CBMIR performance. Our results show that, overall, for the 2D datasets, foundation models deliver superior performance by a large margin compared to CNNs, with the general-purpose self-supervised model for computational pathology (UNI) providing the best overall performance across all datasets and image sizes. For 3D datasets, CNNs and foundation models deliver more competitive performance, with contrastive learning from captions for histopathology model (CONCH) achieving the best overall performance. Moreover, our findings confirm that while using larger image sizes (especially for 2D datasets) yields slightly better performance, competitive CBMIR performance can still be achieved even with smaller image sizes. Our codes to reproduce the results are available at: https://***/masih4/MedImageRetrieval. © 2024, CC BY-NC-ND.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Natural Language Processing and Optimization Approach for Automated Reviewer Assignment

SSRN

引用

SSRN 2024年

作者： Aksoy, Meltem Yanik, Seda Amasyali, Mehmet Fatih Department of Industrial Engineering Istanbul Technical University Sarıyer Istanbul Maslak34485 Turkey Department of Computer Science Research Center Trustworthy Data Science and Security Technical University Dortmund Joseph-vonFraunhofer-Straße 25 Dortmund44227 Germany Department of Computer Engineering Yildiz Technical University Davutpaşa Istanbul Esenler34420 Turkey

Reviewer assignment is a crucial task that involves selecting and assigning qualified reviewers to evaluate projects or paper proposals. The assignment should be based on the expertise of reviewers and the content of proposals to ensure accuracy and fairness of evaluation. This task is typically performed manually, which leads to subjective judgments and requires a significant amount of time and effort. Addressing the inefficiency of traditional manual reviewer assignments, this study introduces a decision support system that automates the assignment task. Utilizing information retrieval, natural language processing, and optimization techniques, it tackles the reviewer assignment problem through a structured, three-stage approach. Initially, it gathers diverse information from various sources to build a comprehensive database of proposals and reviewers. Then, it applies word embedding methods to convert multilingual proposal and reviewer texts into vector representations, and it uses the cosine similarity metric to determine content similarity between each proposal-reviewer pair. Concurrently, it assesses reviewer competency by analyzing their past evaluation performance and areas of expertise through predefined knowledge rules. In the final stage, a multi-objective integer linear programming model assigns reviewers to proposals, optimizing proposal-reviewer similarity and reviewer competency while avoiding conflicts of interest. Furthermore, it explores a max-min approach to improve outcomes for the least-advantaged proposals. This model is enhanced by two additional models to ensure balanced reviewer workloads. The performance of the system is tested with a real-world dataset from the project proposal evaluation process of a regional development agency. The results show that the proposed system significantly outperforms traditional methods. © 2024, The Authors. All rights reserved.

关键词： Integer programming

来源：评论

学校读者我要写书评

暂无评论

MACCoM: A multiple attention and convolutional cross-mixer framework for detailed 2D biomedical image segmentation

引用

computers in Biology and Medicine 2024年 179卷 108847-108847页

作者： Ejiyi, Chukwuebuka Joseph Qin, Zhen Ejiyi, Makuachukwu Bennedith Ukwuoma, Chiagoziem Ejiyi, Thomas Ugochukwu Muoka, Gladys Wavinya Gyarteng, Emmanuel S.A. Bamisile, Olusola O. Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of China Sichuan Chengdu China Pharmacy Department University of Nigeria Nsukka Enugu State Nsukka Nigeria Department of Pure and Industrial Chemistry University of Nigeria Nsukka Enugu State Nsukka Nigeria School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Sichuan Industrial Internet Intelligent Monitoring and Application Engineering Research Center Chengdu University of Technology Chengdu China

The UNet architecture, which is widely used for biomedical image segmentation, has limitations like blurred feature maps and over- or under-segmented regions. To overcome these limitations, we propose a novel network architecture called MACCoM (Multiple Attention and Convolutional Cross-Mixer) – an end-to-end depthwise encoder-decoder fully convolutional network designed for binary and multi-class biomedical image segmentation built upon deeperUNet. We proposed a multi-scope attention module (MSAM) that allows the model to attend to diverse scale features, preserving fine details and high-level semantic information thus useful at the encoder-decoder connection. As the depth increases, our proposed spatial multi-head attention (SMA) is added to facilitate inter-layer communication and information exchange, enabling the network to effectively capture long-range dependencies and global context. MACCoM is also equipped with a convolutional cross-mixer we proposed to enhance the feature extraction capability of the model. By incorporating these modules, we effectively combine semantically similar features and reduce artifacts during the early stages of training. Experimental results on 4 biomedical datasets crafted from 3 datasets of varying modalities consistently demonstrate that MACCoM outperforms or matches state-of-the-art baselines in the segmentation tasks. With Breast Ultrasound Image (BUSI), MACCoM recorded 99.06 % Jaccard, 77.58 % Dice, and 93.92 % Accuracy, while recording 99.50 %, 98.44 %, and 99.29 % respectively for Jaccard, Dice, and Accuracy on the Chest X-ray (CXR) images used. The Jaccard, Dice, and Accuracy for the High-Resolution Fundus (HRF) images are 95.77 %, 74.35 %, and 95.95 % respectively. The findings here highlight MACCoM's effectiveness in improving segmentation performance and its valuable potential in image analysis. © 2024 Elsevier Ltd

关键词： Mixers (machinery)

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Decentralized Signature-Based Access Control

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Han, Jinguang Chen, Liqun Susilo, Willy Southeast University School of Cyber Science and Engineering Jiangsu Nanjing210096 China Ministry of Education Engineering Research Center of Blockchain Application China University of Surrey Department of Computer Science GuildfordGU2 7XH United Kingdom University of Wollongong School of Computing & Information Technology New South Walse WollongongNSW2522 Australia

Decentralized access control reduces trust on the central authority. Considering privacy issues, privacy-preserving decentralized access control (PPDAC) schemes were proposed. However, existing schemes did not consider the unforgeability and aggregtability of access permissions and the traceability of illegal users. In this paper, we propose a privacy-preserving decentralized signature-based access control (PPDSBAC) scheme to fix the above problems. In our scheme, each access permission is a multi- signatures on a keyword, time period and pseudonym, hence is unforgeable and aggregatable. Services are encrypted by using different keywords and time periods. A user can use a pseudonym to access a service directly if he/she gets required access permissions from required authorization authorities. Furthermore, in the case that trace is required, a trace authority can de-anonymise users and identify their real identities. We formalize both the definition and security model of our scheme, and propose a concrete construction. Then, the implementation and evaluation are conducted to show its efficiency. Finally, the security of our scheme is formally proven. The novelty is to construct a PPDAC scheme with unlinkable access requests, unforgeable and aggregatable access permissions, non-interactive authentication and traceability. © 2004-2012 IEEE.

关键词： Authentication

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：