检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Qiu, Zhongxi Hu, Yan Li, Heng Liu, Jiang Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China

Segmentation is vital for ophthalmology image analysis. But its various modal images hinder most of the existing segmentation algorithms applications, as they rely on training based on a large number of labels or hold weak generalization ability. Based on Segment Anything (SAM), we propose a simple but effective learnable prompt layer suitable for multiple target segmentation in ophthalmology multimodal images, named Learnable Ophthalmology Segment Anything (SAM). The learnable prompt layer learns medical prior knowledge from each transformer layer. During training, we only train the prompt layer and task head based on a one-shot mechanism. We demonstrate the effectiveness of our thought based on four medical segmentation tasks based on nine publicly available datasets. Moreover, we only provide a new improvement thought for applying the existing fundamental CV models in the medical field. Our codes are available at website. © 2023, CC BY-NC-SA.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

Supersymmetric Microring Laser Arrays in Higher-Dimensional Space

Supersymmetric Microring Laser Arrays in Higher-Dimensional ...

引用

CLEO: science and Innovations, S and I 2022

作者： Qiao, Xingdu Midya, Bikashkali Gao, Zihe Zhang, Zhifeng Zhao, Haoqi Wu, Tianwei Yim, Jieun Agarwal, Ritesh Litchinitser, Natalia M. Feng, Liang Department of Electrical and Systems Engineering University of Pennsylvania PhiladelphiaPA19104 United States Department of Materials Science and Engineering University of Pennsylvania PhiladelphiaPA19104 United States Department of Electrical and Computer Engineering Duke University DurhamNC27708 United States

To achieve single fundamental supermode operation in coupled arrays, a two-dimensional supersymmetry formalism for mode control is developed and experimentally demonstrated with microring laser arrays. © Optica P...

ISBN: (纸本)9781557528209

关键词：

来源：评论

学校读者我要写书评

暂无评论

On moment relaxations for linear state feedback controller synthesis with non-convex quadratic costs and constraints

On moment relaxations for linear state feedback controller s...

引用

IEEE Conference on Decision and control

作者： Dennis Gramlich Sheng Gao Hao Zhang Carsten W. Scherer Christian Ebenbauer Chair of Intelligent Control Systems RWTH Aachen University Aachen Germany Department of Control Science and Engineering College of Electronic and Information Engineering Tongji University Shanghai P. R. China Chair of Mathematical Systems Theory University of Stuttgart Stuttgart Germany

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

We present a simple and effective way to account for non-convex costs and constraints in state feedback synthesis, and an interpretation for the variables in which state feedback synthesis is typically convex. We achieve this by deriving the controller design using moment matrices of state and input. It turns out that this approach allows the consideration of non-convex constraints by relaxing them as constraints on the expected value of quadratic functions, and that the variables in which state feedback synthesis is typically convexified can be identified with blocks of these moment matrices. The employed relaxation can lead to constraint violation along sample paths.

关键词： State feedback Costs Additives Noise Quadratic programming Time-varying systems

来源：评论

学校读者我要写书评

暂无评论

A Deep Features Based Approach Using Modified ResNet50 and Gradient Boosting for Visual Sentiments Classification

arXiv

引用

arXiv 2024年

作者： Arslan, Muhammad Mubeen, Muhammad Akram, Arslan Abbasi, Saadullah Farooq Ali, Muhammad Salman Tariq, Muhammad Usman Department of Computer Science Loyola University Chicago Chicago United States Department of Computer Science University Of the People CA United States Facultty of Computer Science and Information Technology The Superior University Lahore Lahore Pakistan Department of Electronic Electrical and Systems Engineering University of Birmingham Birmingham United Kingdom

The versatile nature of Visual Sentiment Analysis (VSA) is one reason for its rising profile. It isn't easy to efficiently manage social media data with visual information since previous research has concentrated on Sentiment Analysis (SA) of single modalities, like textual. In addition, most visual sentiment studies need to adequately classify sentiment because they are mainly focused on simply merging modal attributes without investigating their intricate relationships. This prompted the suggestion of developing a fusion of deep learning and machine learning algorithms. In this research, a deep feature-based method for multiclass classification has been used to extract deep features from modified ResNet50. Furthermore, gradient boosting algorithm has been used to classify photos containing emotional content. The approach is thoroughly evaluated on two benchmarked datasets, CrowdFlower and GAPED. Finally, cutting-edge deep learning and machine learning models were used to compare the proposed strategy. When compared to state-of-the-art approaches, the proposed method demonstrates exceptional performance on the datasets presented. © 2024, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Hybrid Visual Verification

Hybrid Visual Verification

引用

International Conference on Confluence The Next Generation Information Technology Summit (Confluence)

作者： S.Nikkath Bushra A. Syed Ismail P. Sasigresa Edwin Singh C Department of Computing Technologies SRM Institute of Scienece and Technology KattankulathurChengalpattu India Department of Data Science and Business Systems SRM Institute of Scienece and Technology KattankulathurChengalpattu India Department: Computer Science and Engineering Veltech University Department: Computer Science and Engineering Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology

Data security is becoming increasingly important as cloud computing advances. Data security is the fundamental problem of all distributed computing systems. Cloud computing enables access to distributed applications and services belonging to various organizations spread across different locations. Cloud computing raises security concerns, because it shares data with distributed users via network in an open terrain. In this system, some of the most important security services like image-based security are handled by the cloud computing system. The proposed scheme takes care of several attacks like botnets, malware and guessing attacks. This scheme is contrasted withseveral established schemesand a new form of authentication is used for the purpose of storing files in a secure manner.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Development of Efficient Wireless Sensor Network for IoT Applications 3

Development of Efficient Wireless Sensor Network for IoT App...

引用

3rd International Conference on Pervasive Computing and Social Networking, ICPCSN 2023

作者： Shankar, S. Deepika, G. Devi, G. Ramesh, S. Srivastava, Suraj Kumar, S Sandeep Sri Krishna Arts and Science College Department of Electronics and Communication Systems Tamil Nadu Coimbatore641008 India St. Peter's Engineering College Department of Electronics and Communication Engineering Telangana Hyderabad500100 India Psna College of Engineering and Technology Department of Electrical and Electronics Engineering Tamil Nadu Dindigul624622 India Institute of Artificial Intelligence and Machine Learning Saveetha University Depattment of Computational Intelligence Tamil Nadu Chennai602105 India I.K. Gujral Punjab Technical University Department of Computer Science and Engineering Mohali Campus-1 Punjab Sahibzada Ajit Singh Nagar160055 India Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering Andhra Pradesh Vaddeswaram522502 India

ISBN: (纸本)9798350322842

This research study provides an overview of the effort that went into developing reliable 'Wireless sensor networks' for IoT applications, including both theoretical study and practical testing. The report also includes the findings of an analysis of variance and regression analysis, which pinpoint the significance of such aspects as network architecture, traffic load, and ambient conditions on the performance of Wireless Sensor Network (WSN). Finally, study provides a discussion on the future research directions, as well as the design and implementation of a Wireless Sensor Network (WSN) for Internet of Things (IoT) applications. © 2023 IEEE.

关键词： Electric power utilization

来源：评论

学校读者我要写书评

暂无评论

UAV -Based Object Tracking Using Siamese Networks and Kalman Filtering

UAV -Based Object Tracking Using Siamese Networks and Kalman...

引用

Multi-Agent systems for Collaborative Intelligence (ICMSCI), International Conference on

作者： S.V Ramanan M. Rajasekar G. Shanmugapriya Sujitha R Akshya. J Mani Deepak Choudhry Department of Electronics and Communication Engineering P PG Institute of Technology Coimbatore Department of Computer Science and Engineering VSB college of Engineering Technical Campus Coimbatore Department of Computer Science and Business Systems Dr.N G.P Institute of Technology Coimbatore Department of Information Technology Christ The King Engineering College Coimbatore Department of Computational Intelligence SRM Institute of Science and Technology Chennai Department of Computing Technologies SRM Institute of Science and Technology Chennai

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509835

The ability to match features and keep track of objects in changing dynamic environments is still an important problem, particularly due to varying noise levels, diverse datasets, and high dimensional feature spaces. Euclidean Distance and Cosine Similarity based matching are existing methods that commonly do not generalize well, particularly in the case of cross domain scenarios, where noise and motion dynamics are powerful factors to limit the accuracy of tracking. However, Deep Metric Learning approaches can extract robust features at the cost of computational efficiency and are unsuitable for real-time applications. This research proposes a Siamese Network framework to solve these problems by using robust feature matching and multi-object tracking. Based on that, the proposed framework uses a dual-branch network architecture in which feature embeddings extracted from paired inputs are mapped through the same shared latent space. Furthermore, the adaptive learning rate leads to efficient convergence during the training. Experimental results show that the Siamese Network not only achieves better accuracy than the existing methods and the more such as Deep Metric Learning but also performs significantly more efficient computation. With an accuracy of 90.2 % at a precision of 88.5%, the framework's inference time is much faster than that of Deep Metric Learning at 2.0ms. Furthermore, the proposed method performs well on the generalization over datasets with a 15% improvement of cross-domain matching accuracy on Euclidean Distance. By striking this balance between accuracy, efficiency and adaptability, it shows its suitability for dynamic and noisy environments for real-time tracking. The proposed method fills existing gaps and achieves a new state of the art for feature matching and tracking performance.

关键词： Accuracy Dynamics Euclidean distance Feature extraction Real-time systems Robustness Object tracking Noise measurement Vehicle dynamics Noise level

来源：评论

学校读者我要写书评

暂无评论

Data-Filtered Prediction with Decomposition and Amplitude-Aware Permutation Entropy for Workload and Resource Utilization in Cloud Data Centers

引用

IEEE Internet of Things Journal 2025年第12期12卷 19189-19201页

作者： Yuan, Haitao Hu, Qinglong Wang, Meijia Wang, Shen Bi, Jing Buyya, Rajkumar Shi, Shuyuan Yang, Jinhong Zhang, Jia Zhou, MengChu Beihang University School of Automation Science and Electrical Engineering Beijing100191 China City University of Hong Kong Department of Computer Science Hong Kong999077 Hong Kong Beijing University of Technology School of Software Engineering The Faculty of Information Technology Beijing100124 China Lab School of Computing and Information Systems MelbourneVIC3010 Australia Beihang University School of Integrated Circuit Science and Engineering Beijing China CSSC Systems Engineering Research Institute Beijing China Southern Methodist University Department of Computer Science The Lyle School of Engineering DallasTX75205 United States New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ07102 United States

In recent years, cloud computing has witnessed widespread applications across numerous organizations. Predicting workload and computing resource data can facilitate proactive service operation management, leading to substantial improvements in quality of service and cost efficiency. However, these data often exhibit non-linearity, high volatility, and interdependencies across different categories, presenting challenges for accurate forecasting. Consequently, there is a critical need to develop a method that thoroughly and comprehensively analyzes all available data to forecast future trends effectively. This work proposes a novel integrated data-enhanced prediction model named SVAPI for achieving high-accuracy workload prediction in cloud computing systems. SVAPI employs the Savitzky-Golay filter, Variational mode decomposition, and the mode selection based on Amplitude-aware Permutation entropy for feature processing, whose features are subsequently utilized by Informer for multivariate joint analysis of the enhanced data, achieving high-precision prediction. Ablation and comparative experiments with advanced prediction models are conducted on the Google cluster trace and other typical datasets. Realistic data-driven results indicate that SVAPI improves the prediction accuracy by 37.7% compared to the original Informer, with each module contributing to the performance enhancement. Furthermore, compared with Autoformer, SVAPI enhances the prediction accuracy of workload, CPU, and memory by 65.6%, 66.9%, and 70.8%, respectively, demonstrating that SVAPI owns strong abilities in noise filtering, feature processing, and multivariate joint analysis for achieving higher prediction accuracy. © 2014 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Thermal Digital Twin of a Multi-Domain System for Discovering Mechanical Faulty Behaviors

Thermal Digital Twin of a Multi-Domain System for Discoverin...

引用

IEEE International Conference on Industrial Informatics (INDIN)

作者： Francesco Tosoni Nicola Dall’Ora Enrico Fraccaroli Sara Vinco Franco Fummi Department of Engineering for Innovation Medicine University of Verona Italy Department of Computer Science University of North Carolina at Chapel Hill USA Department of Control and Computer Engineering Politecnico di Torino Italy

Constructing a holistic digital twin of a system composed of multiple physical domains is crucial for various tasks. In particular, when the simulation is extended with faults, it becomes a very important resource to achieve robust functional safety analysis. This article proposes a new methodology to build non-electrical fault models for the thermal domain. Such thermal faults are defined through an electrical circuit representing the thermal behavior of the system, known as the Cauer network, based on the physical analogies between the two domains. Including this thermal representation in a multi-domain system allows to simulate the interconnections between different physical domains, thus achieving a more realistic system behavior and evaluating the mutual impact of different domains (e.g., mechanical, electrical and thermal). The entire methodology is applied to a complex case of study implemented by using Verilog-AMS as a proof of concept.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Designing a Medication Assistive Intelligent System with Automated Insulin-dose Prediction for Elderlies

Towards Designing a Medication Assistive Intelligent System ...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Abrar Ahmed Minhazul Islam Mahi M. Akhtaruzzaman Jamal Uddin Tanvin Fahim Shahryer Sachitra Halder Muhammad Towfiqur Rahman Md Shofiqul Islam Department of Computer Science and Engineering Military Institute of Science and Technology Dhaka Bangladesh Quantum Robotics and Automation Research Group (QRARG) Dhaka Bangladesh Department of Computer Science & Engineering University Asia Pacific (UAP) Dhaka Bangladesh Institute for Intelligent Systems Research and Innovation (IISRI) Deakin University Geelong Victoria

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Medication management poses significant challenges for many patients, particularly the elderlies, who often struggle with keeping track of their medication schedules and taking the correct dosages. To address this issue, this study aims to design an AI-integrated Medication Management and Assistive Unit (AMMAU) for elderlies with some important features like, schedule reminder, medicine recognition and count, early indication of shortage, sorted slots, and so on. Automated insulin-dose prediction and alert system makes the system unique and more demanding at these current scenarios. For this study, a suitable machine learning model is designed, analyzed, verified, and embedded in the proposed system so that the elderly diabetic patients can get alert, further compare with the current insulin-doses. Though, the proposed system is currently focusing on only the Basal insulin doses prediction, the system will definitely reduce the risks associated with wrong management of medication by degrading the chances of missed doses or taking wrong pills for elderlies at home.

关键词： Training Schedules Analytical models Prototypes Machine learning Predictive models Internet of Things Older adults Insulin Physical design

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：