检索结果-内蒙古大学图书馆

Deep Learning-Based Mask Identification System Using ResNet Transfer Learning Architecture

computer Systems Science & engineering 2024年第2期48卷 341-362页

作者： Arpit Jain Nageswara Rao Moparthi A.Swathi Yogesh Kumar Sharma Nitin Mittal Ahmed Alhussen Zamil S.Alzamil MohdAnul Haq Department of Computer Science and Engineering KoneruLakshmaiah Education FoundationVaddeswaram522302India Department of Computer Science and Engineering Sreyas Institute of Engineering and TechnologyHyderabad500068India University Centre for Research and Development Chandigarh UniversityMohali140413India Department of Computer Engineering College of Computer and Information SciencesMajmaah UniversityAl-Majmaah11952Saudi Arabia Department of Computer Science College of Computer and Information SciencesMajmaah UniversityAl-Majmaah11952Saudi Arabia

Recently,the coronavirus disease 2019 has shown excellent attention in the global community regarding health and the *** Health Organization(WHO)and many others advised controlling Corona Virus Disease in *** limited treatment resources,medical resources,and unawareness of immunity is an essential horizon to *** all resources,wearing a mask is the primary non-pharmaceutical intervention to stop the spreading of the virus caused by Severe Acute Respiratory Syndrome Coronavirus 2(SARS-CoV-2)*** countries made masks mandatory to prevent *** such enforcement,automatic and effective face detection systems are *** study presents a face mask identification approach for static photos and real-time movies that distinguishes between images with and without *** contribute to society,we worked on mask detection of an individual to adhere to the rule and provide awareness to the public or *** paper aims to get detection accuracy using transfer learning from Residual Neural Network 50(ResNet-50)architecture and works on detection *** experiment is tested with other popular pre-trained models such as Deep Convolutional Neural Networks(AlexNet),Residual Neural Networks(ResNet),and Visual Geometry Group Networks(VGG-Net)advanced *** proposed system generates an accuracy of 98.4%when modeled using Residual Neural Network 50(ResNet-50).Also,the precision and recall values are proved as better when compared to the existing *** outstanding work also can be used in video surveillance applications.

关键词： Transfer learning depth analysis convolutional neural networks(CNN) COVID-19

来源：评论

学校读者我要写书评

暂无评论

A Simple but Effective Approach for Unsupervised Few-Shot Graph Classification 24

A Simple but Effective Approach for Unsupervised Few-Shot Gr...

引用

33rd ACM Web Conference, WWW 2024

作者： Liu, Yonghao Huang, Lan Cao, Bowen Li, Ximing Giunchiglia, Fausto Feng, Xiaoyue Guan, Renchu College of Computer Science and Technology Jilin University Changchun China Department of Information Engineering and Computer Science University of Trento Trento Italy Key Laboratory of Symbolic Computation and Knowledge Engineering of the Ministry of Education China

ISBN: (纸本)9798400701719

Graphs, as a fundamental data structure, have proven efficacy in modeling complex relationships between objects and are therefore found in wide web applications. Graph classification is an essential task in graph data analysis, which can effectively assist in extracting information and mining content from the web. Recently, few-shot graph classification, a more realistic and challenging task, has garnered great research interest. Existing few-shot graph classification models are all supervised, assuming abundant labeled data in base classes for meta-training. However, sufficient annotation is often challenging to obtain in practice due to high costs or demand for expertise. Moreover, they commonly adopt complicated meta-learning algorithms via episodic training to transfer prior knowledge from base classes. To break free from these constraints, in this paper, we propose a simple yet effective approach named SMART for unsupervised few-shot graph classification without using any labeled data. SMART employs transfer learning philosophy instead of the previously prevailing meta-learning paradigm, avoiding the need for sophisticated meta-learning algorithms. Additionally, we adopt a novel mixup strategy to augment the original graph data and leverage unsupervised pretraining on these data to obtain the expressive graph encoder. We also utilize the prompt tuning technique to alleviate the overfitting and low fine-tuning efficiency caused by the limited support samples of novel classes. Extensive experimental results demonstrate the superiority of our proposed approach, significantly surpassing even leading supervised few-shot graph classification models. Our code is available here. © 2024 ACM.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Joint Optimization of PAoI and Queue Backlog with Energy Constraints in LoRa Gateway Systems 19th

Joint Optimization of PAoI and Queue Backlog with Energy ...

引用

19th EAI International Conference on Collaborative Computing: Networking, Applications and Worksharing, CollaborateCom 2023

作者： Shi, Lei Ji, Rui Wei, Zhen Feng, Shilong Li, Zhehao School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China Engineering Research Center of Safety Critical Industrial Measurement and Control Technology Ministry of Education Hefei230009 China

ISBN: (纸本)9783031545306

Peak Age of information(PAoI), as a performance indicator representing the freshness of information, has attracted the attention of researchers in recent years. The data packet transmission rate in the LoRa network determines the information freshness level for system packets. In order to study the optimal scheduling of data packets, we try to use the PAoI to describe the real-time level of the end devices(EDs) and reduce it. We use edge servers to process monitoring data packets at the edge of the network to improve the efficiency of EDs and the information freshness level of data. Since packet transmission will be constrained by EDs battery queue energy and gateway queue backlog, we propose an optimization problem that aims to minimize the long-term average PAoI of EDs while ensuring network stability. With the Lyapunov optimization framework, the long-term stochastic optimization problem is transformed into a single-slot optimization problem. Furthermore, to avoid the problem of too large search space, we propose a dynamic strategy space reduction algorithm (SSDR) to shrink the strategy space. The simulation experiments show that our SSDR algorithm can optimize the PAoI index of EDs in various situations and satisfy the constraints of long-term optimization. © ICST Institute for computer Sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Secured Authentication Scheme for E-Healthcare in SIoT Using ECC and Improved Hash Function 4th

Secured Authentication Scheme for E-Healthcare in SIoT Using...

引用

4th International Conference on Frontiers in Computing and Systems, COMSYS 2023

作者： Arpitha, T. Chouhan, Dharamendra Shreyas, J. Department of Computer Science and Engineering University Visvesvaraya College of Engineering Bangalore University Bengaluru India Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India

ISBN: (纸本)9789819726134

The convergence of Smart Internet of Things (IoT) and security is a new study topic. In this context, a new IoT trend known as Social IoT (SIoT) was examined for this study. The Social IoT has the advantage of connectivity demonstrated within the network of connected things via the IoT. In Social IoT, a wide range of IoT devices, such as wearables or monitoring sensors, are utilized in e-healthcare applications and the risk of identity theft and impersonation is a significant concern. To address this challenge, we propose a scheme that incorporates multi-factor authentication, elliptic curve cryptography (ECC), and hash functions scheme to provide a robust solution for secure and privacy-preserving authentication. Our system provides privacy protection along with mutual authentication and session key negotiation at a considerably low computational cost (of up to 71.04% compared to other existing techniques). © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Automatic skin lesion classification using a novel densely connected convolutional network integrated with an attention module

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第7期14卷 8943-8956页

作者： Shan, Pufang Chen, Jialei Fu, Chong Cao, Lin Tie, Ming Sham, Chiu-Wing School of Computer Science and Engineering Northeastern University Shenyang110819 China Engineering Research Center of Security Technology of Complex Network System Ministry of Education Shenyang China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang110819 China School of Information and Communication Engineering Beijing Information Science and Technology University Beijing100101 China Science and Technology on Space Physics Laboratory Beijing100076 China School of Computer Science The University of Auckland Auckland New Zealand

Automated skin lesion classification in dermoscopy images remains challenging due to the existence of artefacts and intrinsic cutaneous features, diversity of lesion morphology, insufficiency of training data, and class imbalance problem. To address these challenges, we propose a new densely connected convolutional network termed AttDenseNet-121, which is obtained by integrating the convolutional block attention module (CBAM) into DenseNet-121. CBAM is a simple yet efficient attention module, which is further improved by adding a novel pooling layer (multiscale-pooling) to effectively enhance its attention capability. The optimized CBAM strengthens the representation power of DenseNet-121 by emphasizing meaningful features and suppressing unnecessary ones, thus significantly enhancing the classification performance. Besides, to handle the imbalanced dataset, we employ an improved focal loss function rather than the traditional cross-entropy loss function to train AttDenseNet-121. The improved focal loss is calculated from a different perspective compared with the original focal loss. It makes the distribution of positive samples and negative samples in each batch the same as that in the original data, successfully mitigating the negative influence of class imbalance on this multi-class classification task. We conduct extensive experiments on the public benchmark dataset (HAM10000) and the results indicate that our proposed method achieves superior performance compared with that of the baselines and state-of-the-art algorithms. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Roadside IRS Assisted Task Offloading in Vehicular Edge Computing Network 19th

Roadside IRS Assisted Task Offloading in Vehicular Edge Com...

引用

19th EAI International Conference on Collaborative Computing: Networking, Applications and Worksharing, CollaborateCom 2023

作者： Xie, Yibin Shi, Lei Li, Zhehao Ding, Xu Liu, Feng School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China Engineering Research Center of Safety Critical Industrial Measurement and Control Technology Ministry of Education Hefei230009 China

ISBN: (纸本)9783031545207

Vehicular edge computing (VEC) has been recognized as a promising technique to process delay-sensitive vehicular applications. Nevertheless, in order to accommodate the rapid growth in the number of connected vehicles, it’s inevitable that there will be an increasing deployment of conventional infrastructure with limited communication ranges. This could potentially lead to escalating costs and impede the full realization of the VEC system. In this paper, a roadside intelligent reflecting surface (IRS) assisted VEC network is introduced, where the IRS is deployed outside the coverage of roadside units (RSUs) to extend the service range. Furthermore, the maximum total number of successful offloading tasks problem within the scheduling time problem is formulated, encompassing the optimization of offloading decisions, computation resource allocation and phase shift of IRS. To tackle the formulated challenging problem, we first decouple the original problem into two subproblems. Then, a heuristic algorithm is proposed, where a many-to-one matching algorithm is proposed to joint optimize offloading decision and the computation resource, and an iterative algorithm is utilized to optimize the phase shift coefficients of IRS. The simulation results validate the effectiveness of the proposed algorithm in comparison to other schemes, and the IRS can effectively maintain network performance even when there are intervals in RSU coverage areas. © ICST Institute for computer Sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Learner Assessment System in e-Learning with OBE Approach: Activity Performance, Ability Level and Recommendation

HighTech and Innovation Journal

引用

HighTech and Innovation Journal 2024年第3期5卷 572-602页

作者： Yuniarti, Wenty D. Hartati, Sri Priyanta, Sigit Surjono, Herman D. Department of Computer Science and Electronics Universitas Gadjah Mada Yogyakarta55281 Indonesia Department of Information Technology Universitas Islam Negeri Walisongo Semarang50185 Indonesia Department of Informatics Engineering Education Faculty of Engineering Universitas Islam Negeri Walisongo Yogyakarta55281 Indonesia

E-learning can lead learners to achieve learning outcomes if it is designed based on several principles. One is applying assessments that motivate and inform ability levels. In Outcome-based education (OBE), assessment is integral to the system. However, e-learning has limitations in providing assessment instruments according to needs, such as assessing complex and detailed aspects and accommodating a variety of numerical and linguistic assessment data. Moreover, the presence and involvement of learners affect their performance and learning outcomes. This study proposes a learner assessment system in e-learning with the OBE approach, including learning design, activity performance analysis, ability level determination, and recommendations. This system adds the e-rubric to e-learning to overcome instrument limitations and accommodate comprehensive assessments. Various numerical and linguistic assessment data are unified using 2-tuple fuzzy linguistics, producing ability levels as two tuples. Performance analysis was based on event log data using descriptive statistical technique and alignment-based conformance checking, from frequency, time, and sequence of activity objects, resulting in five activity performance variables. The performance value of each variable is converted into High, Medium, or Low levels. The ability and performance levels are processed using rule-based methods to produce recommendations for learning stages and activity performance directions. The results of this research can be used as input for academic stakeholders and online learning providers and potentially be applied to the advancement of e-learning in higher education. © Authors retain all copyrights.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Incorporating Artificial Intelligence Powered Immersive Realities to Improve Learning using Virtual Reality (VR) and Augmented Reality (AR) technology 3

Incorporating Artificial Intelligence Powered Immersive Real...

引用

3rd International Conference on Applied Artificial Intelligence and Computing, ICAAIC 2024

作者： Balushi, Jasim Sharki Ghulam Al Jabri, Malak Ibrahim Ahmed Al Palarimath, Suresh Maran, Pyingkodi Thenmozhi, K. Balakumar, C. Univesity of Technology and Applied Sciences College of Computing and Information Sciences Information Technology Department Salalah Oman Kongu Engineering College Department of Computer Applications Erode India Department of Computer Science Bengaluru India Karpagam Academy of Higher Education Department of Computer Science Coimbatore India

ISBN: (纸本)9798350375190

This study analyzes the transformative potential of combining artificial intelligence (AI) with immersive technologies, specifically virtual reality (VR) and augmented reality (AR), to elevate educational outcomes significantly. The synergy of AI and immersive technologies presents a unique opportunity to enhance students' motivation, memory retention, and comprehension of complex subjects, creating a distinct and impactful learning experience. The research comprehensively explores the current understanding of the effectiveness of VR and AR in educational settings, placing special emphasis on their applications in diverse fields such as Science, technology, engineering, And Mathematics (STEM), medicine, language acquisition, and the development of interpersonal skills. Beyond highlighting the immense potential, the study acknowledges and addresses the obstacles that include the costs associated with equipment, training, and content creation. It underscores the importance of collaborative efforts among technology suppliers, educators, and content producers to overcome these challenges effectively. By recognizing and navigating these hurdles, the study aims to provide practical insights for educational institutions and policymakers, advocating for the integration of AI-powered immersive technology to enhance the overall learning experience for students. The findings underscore the transformative impact of such technologies, emphasizing their role in shaping the future of education and preparing students for the dynamic challenges of the modern world. © 2024 IEEE.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Handwriting Detection Using Object Detection Models 4

Handwriting Detection Using Object Detection Models

引用

4th International Conference on Advancement in Electronics and Communication engineering, AECE 2024

作者： Raj, Mayank Chhabra, Prachi Shrivastava, Aparna Singh, Vikas Kumar, Abhishek Sharma, Palak Jss Academy of Technical Education Department of Information Technology Noida India School of Computer Science & Engineering Iilm University Greater Noida India Jss Academy of Technical Education Department of Computer Science & Datascience Noida India

ISBN: (纸本)9798350364729

In this paper, we propose a new method for handwriting text detection using the YOLOV5 object detection neural network. Our approach accurately identifies text sequences regardless of orientation, size, or type (handwritten or printed) and locates handwritten text regions within images. We then utilize the TrOCR model by Hugging Face, which employs an encoder-decoder mechanism to extract and digitize text from the images. To train and evaluate our model, we introduce a clinical receipts dataset tailored for handwriting text detection, addressing challenges like variations in styles, sizes, and orientations. Through extensive experimentation, we demonstrate our approach's effectiveness on benchmark datasets, achieving competitive performance against existing methods. Additionally, we explore potential applications of handwriting text detection in tasks such as OCR, document classification, and information extraction, presenting a promising solution for robust and efficient handwriting text detection in diverse real-world scenarios. © 2024 IEEE.

关键词： HTR Hugging Face OCR TrOCR YOLOV5

来源：评论

学校读者我要写书评

暂无评论

Model-driven Time-varying Channel Estimation Method for OFDM System Based on Neural Network 24

Model-driven Time-varying Channel Estimation Method for OFDM...

引用

24th IEEE International Conference on Communication technology, ICCT 2024

作者： Zhu, Yue Fan, Dian Mu, Yunping Wang, Gongpu China Academy of Information and Communications Technology Beijing100191 China Beijing Jiaotong University Engineering Research Center of Network Management Technology for High Speed Railway Ministry of Education School of Computer and Information Technology Beijing100044 China

ISBN: (纸本)9798350363760

Channel estimation plays an important role in orthogonal frequency division multiplexing (OFDM) systems, which directly affects the performance of wireless communication systems in high-speed mobile scenarios. In this paper, we propose a model-driven channel estimation method for OFDM systems with doubly selective channels. Instead of completely abandoning traditional communication modules, we first obtain rough channel estimation using the least square (LS) method and then use a neural network called RefineNet for accuracy improvement. The results show that our proposed method significantly outperforms the linear minimum mean squared error (LMMSE) method. © 2024 IEEE.

关键词： Orthogonal frequency division multiplexing

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：