检索结果-内蒙古大学图书馆

2025 International Conference on Ambient Intelligence in Health Care, ICAIHC 2025

作者： Prajapat, Gopal Krishan Yadav, Dharmendra Bikaner Technical University Dept. of Computer Science and Engineering Bikaner India

ISBN: (纸本)9798331530259

Computed Tomography (CT) is essential in medical diagnostics, providing detailed images of internal structures. Noise and blur introduced during image acquisition can com-promise diagnostic accuracy, especially in low-dose CT (LDCT) scans aimed at reducing radiation exposure. Deep learning methods have emerged as effective solutions for addressing these challenges, with significant advancements in both denoising and deblurring techniques. This paper reviews recent deep learning approaches for CT image enhancement, covering supervised and self-supervised methods. The paper also explores emerging trends such as diffusion models and transformer-based architectures that offer improved image quality and restoration efficiency. Simultaneous denoising and deblurring techniques are examined for their potential in clinical applications. Promising future directions include the use of large-scale pretrained models, zero-shot learning, and prompt-based frameworks, which have the potential to further enhance CT image quality while maintaining low radiation exposure, thereby advancing diagnostic capabilities. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Navigating Breakthroughs in Pediatric Brain Stroke Detection: A Comprehensive Survey

Navigating Breakthroughs in Pediatric Brain Stroke Detection...

引用

Knowledge engineering and Communication Systems (ICKECS), International Conference on

作者： N R Deepak Santosh Kumar Paital Saritha Verma Tenzin Choeyang Udit Kumar Singh Bhavya dept. of Computer Science and Engineering HKBK College of Engineering Bengaluru India

ISBN: (数字)9798350359688

ISBN: (纸本)9798350359695

Brain strokes, characterized by sudden interruptions in cerebral blood flow, pose a significant health concern, especially in children, where detection is intrinsically challenging due to limitations in existing methodologies. Recent attempts leveraging Machine Learning with Deep Learning techniques have grappled with accuracy and efficiency shortcomings in addressing this critical issue. In response, our innovative model integrates the Extreme Learning Machine (ELM), Convolutional Neural Networks (CNNs), and Generative Adversarial Networks (GANs) algorithms. Rigorously evaluated using a comprehensive dataset, our ELM + CNN + GAN model demonstrates unparalleled accuracy, precision, and recall rates, surpassing benchmarks set by conventional approaches like the Random Forest algorithm and the AlexNet + SVM hybrid model. Notably, our model excels in diagnostic speed, offering a promising avenue for swift and efficient early detection of pediatric brain strokes. The strategic incorporation of ELM, recognized for its efficiency in single layer neural networks, along with the analytical power of CNN for image analysis and GAN for synthetic data generation, establishes a robust and comprehensive model. By outperforming existing methods, our approach signifies a substantial leap forward in addressing this critical health concern, revolutionizing stroke detection with enhanced accuracy and speed.

关键词： Surveys Support vector machines Technological innovation Accuracy Stroke (medical condition) Brain modeling Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

VisualBlend - A Mixed Approach on Forensic Image Generation and Recognition

VisualBlend - A Mixed Approach on Forensic Image Generation ...

引用

IEEE Recent Advances in Intelligent Computational Systems (RAICS)

作者： Pillai Midhun Unnikrishna Vinayak Pulkuzhiyil Jyothika Krishna Naveen Wilson Veena A. Kumar Dept. of Computer Science and Engineering Saintgits College Of Engineering Kottayam India

ISBN: (数字)9798350381689

ISBN: (纸本)9798350381696

An extensive use case of Generative Adversarial Networks (GANs) that not only combines artistic creativity with technological innovation but also significantly improves identification precision, filling in critical gaps in the smooth integration of textual descriptions with visual representations. The motivation is rooted on the difficulties that currently exist in accurately identifying human face sketches by directly transforming them into images, an area where prior approaches were inadequate. Our special dual-function technology is designed to get beyond these obstacles and provides a fresh approach for forensic investigators, artists, and sculptors. It has the ability to convert text to image as well as sketch to image. This easy transformation of textual and drawing data represents an integrated combination of creativity and technology that is set to have an impact on a number of disciplines, such as suspect identification and crime scene reconstruction.

关键词： Visualization Technological innovation Law enforcement Text recognition Forensics Face recognition Computational modeling Text to image Streaming media Creativity

来源：评论

学校读者我要写书评

暂无评论

Using Machine Learning Algorithms in Edge Intelligence for Ultra-Reliable and Low-Latency Communications in the Industrial Metaverse

Using Machine Learning Algorithms in Edge Intelligence for U...

引用

International CSI computer Conference, CSICC

作者： Reza Mohammadvand Naser Mozayani dept. Computer Engineering Iran University of Science and Technology Tehran Iran

ISBN: (数字)9798331523114

ISBN: (纸本)9798331523121

Given the rapid advancements in technologies such as the metaverse and digital twins, the need for effective integration of communication, computation, and storage in complex systems and edge computing environments is becoming increasingly apparent. This paper proposes an architecture to support an industrial metaverse based on digital twins that optimizes resources by leveraging mobile edge computing and ultra-reliable low-latency communication. This architecture reduces latency by employing task offloading and storage methods on nearby edge servers, meeting the precise requirements of future metaverses in terms of reliability and latency minimization. The proposed model, utilizing reinforcement learning algorithms such as Deep Q-Network (DQN), Double Deep Q-Network (DDQN), and Dueling Double Deep Q-Network (Dueling DDQN), is capable of achieving adaptability in dynamic conditions and making intelligent decisions. The simulation results demonstrate that the proposed method reduces by more than 10% on average compared to the best method available in the literature and improves resource utilization efficiency.

关键词： Metaverse Heuristic algorithms Computational modeling Simulation computer architecture Artificial neural networks Ultra reliable low latency communication Digital twins Delays Optimization

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Brain Tumor Classification: CNN Models with Saliency Map Insights

Deep Learning for Brain Tumor Classification: CNN Models wit...

引用

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

作者： M Shahriar Mahmud Rafi Aisha Tasnim Aishy Tazrian Alam Mohammed Maruf Hossen Dept. of Computer Science and Engineering East Delta University Chittagong Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

This paper aims to classify brain tumors using Convolutional Neural Networks (CNNs) applied to MRI images. We employed four models: a custom CNN, VGG16, ResNet101, and a hybrid model combining VGG16 and ResNet101, with saliency maps used for model interpretability. The custom CNN achieved the highest accuracy of 98.01%, followed by VGG16 (96.33%), ResNet101 (90%), and the hybrid model (97.10%). Saliency maps provided insights into the features driving predictions, enhancing model transparency. Our findings demonstrate the effectiveness of CNNs for brain tumor classification, with the custom CNN performing the best, and suggest further research into model explainability, data augmentation, and hybrid techniques to improve accuracy and generalizability.

关键词： Deep learning Accuracy Magnetic resonance imaging Brain tumors Transforms Brain modeling Feature extraction Convolutional neural networks Reliability Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

A Graph Based Approach For Detecting Sybil Node In Online Social Networks

A Graph Based Approach For Detecting Sybil Node In Online So...

引用

International Conference on Computing and Networking Technology (ICCNT)

作者： Sucheta Chandra Priyanka Das Dept. of Computer Application and Science Institute of Engineering & Management Kolkata

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Sybil attacks pose a significant threat to the security and integrity of online social networks by allowing malicious actors to create multiple fake identities to manipulate network behavior. The abstract outlines a method for detecting Sybil attacks in online social networks using graph analysis. It involves constructing a graph representation of the network, calculating node centrality measures such as in-degree and out-degree, and setting a degree threshold to identify potentially malicious nodes. Additional parameters like clustering coefficient and community structure are also considered to improve detection accuracy. The proposed approach aims to enhance the security of online social networks by identifying and isolating Sybil identities, thus preserving network integrity and trustworthiness. The authors’ firmly assert that employing graph-based theory for Sybil node detection in online social networks stands out as one of the most effective methods. This approach not only offers robustness but also scalability, allowing for efficient detection even in large-scale networks. With its ability to unveil intricate relationships, anomalies, and patterns within the network, graph-based theory emerges as a cornerstone in the fight against Sybil attacks.

关键词： Accuracy Social networking (online) Scalability Robustness Security

来源：评论

学校读者我要写书评

暂无评论

Evaluating Low-Resource Lane Following Algorithms for Compute-Constrained Automated Vehicles

arXiv

引用

arXiv 2024年

作者： Froemming-Aldanondo, Beñat Rastoskueva, Tatiana Evans, Michael Machado, Marcial Vadella, Anna Johnson, Rickey Escamilla, Luis Jostes, Milan Butani, Devson Kaddis, Ryan Chung, Chan-Jin Siegel, Josh Dept. of Computer Science & Engineering University of Minnesota Minneapolis United States Dept. of Computer Science The University of Arizona Tucson United States Dept. of Computer Science Old Dominion University Norfolk United States Dept. of Computer Science & Engineering The Ohio State University Columbus United States Dept. of Computer Science SWE Butler University Indianapolis United States Dept. of Computer Science North Carolina A&T State University Greensboro United States Dept. of Computer Science New Mexico State University Las Cruces United States Dept. of Math & Computer Science Lawrence Technological University Southfield United States DeepTech Lab Michigan State University East Lansing United States

Reliable lane-following is essential for automated and assisted driving, yet existing solutions often rely on models that require extensive computational resources, limiting their deployment in compute-constrained vehicles. We evaluate five low-resource lane-following algorithms designed for real-time operation on vehicles with limited computing resources. Performance was assessed through simulation and deployment on real drive-by-wire electric vehicles, with evaluation metrics including reliability, comfort, speed, and adaptability. The top-performing methods used unsupervised learning to detect and separate lane lines with processing time under 10 ms per frame, outperforming compute-intensive and poor generalizing deep learning approaches. These approaches demonstrated robustness across lighting conditions, road textures, and lane geometries. The findings highlight the potential for efficient lane detection approaches to enhance the accessibility and reliability of autonomous vehicle technologies. Reducing computing requirements enables lane keeping to be widely deployed in vehicles as part of lower-level automation, including active safety systems. © 2024, CC BY.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Towards Touchless Interaction: Implementing Hand Gesture Recognition for Presentation and Media Control

Towards Touchless Interaction: Implementing Hand Gesture Rec...

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Shailaja Uke Atifa Shaikh Harshada Rayate Anjali Kamble Sanavi Rahane Dept. Computer Science and Engineering Vishwakarma Institute of Technology Pune India

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

The integration of hand gesture recognition with modern multimedia systems offers a novel and intuitive interface for users to interact with. This research implements a reliable framework for gesture-based control of slide presentations and media functionalities, using real-time hand detection and tracking through computer vision technologies such as OpenCV and cvzone. Recognizing individual hand gestures, the system allows for smooth navigation through slides in a presentation, annotating on slides, changing the brightness, managing windows like window minimization and maximization, and controlling media playback. The solution uses machine learning-based hand pose estimation along with optimal gesture mappings to make the recognition highly accurate and responsive. This research is meant to represent the possibility of hands-free interaction in professional and academic settings. The approach also studies applications in remote work scenarios, thereby increasing the incorporation of gesture recognition technology into everyday computing work. Moreover, the system is flexible and scalable and has the potential to integrate with virtual reality systems, remote learning platforms, and interactive kiosks. The proposed system, under controlled conditions, achieves an average accuracy of about 85 percent to 90 percent. It recognizes predefined hand gestures for navigation and interaction. When multiple gestures are shown at the same time, the system might experience temporary lags, but it recovers functionality within a few seconds. Thus, this research underlines the role of gesture recognition technology in transforming the way a user experience is provided.

关键词： Hands Productivity computer vision Technological innovation Accuracy Navigation Gesture recognition Media Real-time systems Windows

来源：评论

学校读者我要写书评

暂无评论

Decentralized Network for Cloud Storage 2

Decentralized Network for Cloud Storage

引用

2nd Edition of IEEE Delhi Section Owned Conference, DELCON 2023

作者： Chirag, N. Kshitij, K. Gupta, Gaurav Sir M Visvesvaraya Institute of Technology Dept. of Information Science Engineering Bangalore India Shoolini University Yogananda School of Artificial Intelligence Computer and Data Sciences H.P Solan India

ISBN: (纸本)9798350322064

The amount of data loss on corporate servers in cloud environments has increased significantly. In the cloud, there are many security compromises and account hijackings, resulting in severe vulnerabilities for the service. Cloud computing is one of the most popular on-demand services. Over the past decade, cloud computing has been incorporated into various industries, including health care and the military. Security has, however, been a major concern with cloud computing. There have been many approaches and security implications implemented. However, when it comes to the security and privacy of data, there are loopholes. By decentralizing the cloud, which is currently a centralized entity, the authors attempt to improve cloud security. It is possible to solve these problems by deploying a decentralized cloud. Client software encrypts, fragments data, and then a client engine distributes the data among a number of hosting nodes (servers) spread throughout the world. As a result, the user has complete ownership of his data. In their study, the authors found that decentralizing cloud solved most of the problems of centralized cloud, since it increases the security of data by fragmenting and encrypting fragments. © 2023 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Glaucoma Insights: a Visual Comparative Analysis of Fundus Images Across Diverse datasets

Glaucoma Insights: a Visual Comparative Analysis of Fundus I...

引用

Smart Systems for applications in Electrical sciences (ICSSES), International Conference on

作者： Chunduri Yoshitha G Vishwa Prakashini Neha S Aiswariya Milan K Dept. of Computer Science and Engineering Amrita School of Computing Bengaluru India

ISBN: (数字)9798331507671

ISBN: (纸本)9798331507688

Glaucoma is a major global health issue that can lead to irreversible vision loss if untreated. Early diagnosis and continuous monitoring are crucial to mitigating its impacts. Fundus images are necessary to evaluate, diagnose glaucoma and offer an in-dept. perception of retinal structures and features indicative of the condition. The project aims to compare the features of fundus images from three widely recognized datasets, they are REFUGE, ODIR, and DRONIS-DB focusing on analyzing critical metrics such as pixel intensity, brightness, image channel analysis, edge mean, and saturation means. Advanced visualizations are employed to ensure a comprehensive analysis of the features. The main objective of the work is to provide a structured data-driven approach for feature extraction and visualization, enhancing our understanding. This project contributes to advancing diagnostic precision, supporting clinical decision-making, and enabling researchers to explore innovative solutions in ophthalmology by bridging the gap between medical imaging data and actionable insights.

关键词： Glaucoma Visualization Decision making data visualization Feature extraction Smart systems Retina Ophthalmology Biomedical imaging Resistance heating

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：