检索结果-内蒙古大学图书馆

AI-Driven Smart Healthcare for Society 5.0

作者： Akshat Bokdia Siri R Kulakarni Nitin Lodha Vijayarajan V V. B. Surya Prasath School of Computer Science and Engineering Vellore Institute of Technology Vellore Vellore India Department of Computer Science University of Cincinnati Cincinnati USA

ISBN: (数字)9798331536336

ISBN: (纸本)9798331536343

In medical informatics, developing efficient image retrieval methods is vital for the research and development of diagnosis and treatment processes. This study evaluates three different feature extraction methods that could be beneficial for enhancing context-based image retrieval (CBIR) in medical imaging applications. The techniques used are: Discrete Wavelet Transform (DWT) with Singular Value Decomposition (SVD), a combination of Sobel operators with DWT and SVD, and Autoencoders. Such hybrid feature extraction techniques are applied to gain salient characteristics and detect trends in the given image data. The BRATS dataset, which consists of various multi-modal MRI scans, is used in the study to validate the efficacy of each technique. To assess the performance of each technique, the InceptionNet V3 model is trained individually on each of these sets of feature-extracted images. Results show that, in terms of robustness and accuracy in classifying medical images, the DWT with SVD method outperforms the others. This research implicates choosing feature extraction techniques appropriately to improve CBIR systems in healthcare.

关键词： Accuracy Magnetic resonance imaging Image retrieval Autoencoders Medical services Transforms Feature extraction Discrete wavelet transforms Medical diagnostic imaging Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

Deep Convolutional Multi-Relational Graph Attention Network for Autonomous Drone Navigation in Urban Wind Zones

Deep Convolutional Multi-Relational Graph Attention Network ...

引用

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

作者： C. Pabitha K. Paul Joshua S Pradeep Sophia S Rajesh Sharma R D. Karthik Prabhu Department of Computer Science and Engineering SRM Valliammai Engineering College Chennai Tamil Nadu India Department of Computer Science and Engineering SCAD College of Engineering and Technology Tirunelveli Tamil Nadu India Department of Computer Science and Engineering MAlla Reddy Engineering College for Women Secunderabad Telangana India Department of computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore Tamil Nadu India Department of Computer Technology Alliance College of Engineering and Design Alliance University Bangalore India Department of Electrical and Electronics Engineering Ramco Institute of Technology Rajapalaiyam Tamil Nadu India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509835

Drones are essential for civil engineering operations like logistics and data collecting. Current autonomous drone studies mainly concerns itself with safe path planning in static scenarios; however one of the major challenges of over urban environment is often dynamic factors such as wind, building layouts, and signal coverage that can significantly impact direction stability. For autonomous navigation across dynamic wind zones, this study proposed a new Shape-aware Mesh Dual Lightweight Deep Convolutional Multi-Relational Graph Attention Network with Growth Optimizer (SAMDLDCMRGANet-GO) that leverages local data from RGB cameras and GPS. The method incorporates a Dual Aggregation Transformer for feature extraction, Shape-aware Mesh Normal Filtering for preprocessing, and the LDCMRGAN et for efficient path planning. The proposed approach demonstrates its efficacy in navigating complicated surroundings while retaining high precision in obstacle avoidance and target acquisition, achieving an impressive 99.14% recall, 99.15% accuracy, an outstanding EVA of 0.99, and a low RMSE of 0.41.

关键词： Navigation Convolution Filtering Urban areas Transformers Feature extraction Cameras Path planning Drones Global Positioning System

来源：评论

学校读者我要写书评

暂无评论

Leveraging Sequential Deep Learning Models for Robust Compiler Bug Detection in Software Logs

Leveraging Sequential Deep Learning Models for Robust Compil...

引用

Intelligent Data Communication Technologies and Internet of Things (IDCIoT), International Conference on

作者： Sahil Mishra Utkarsh Tiwari Kavitha C. R. Department of Computer Science and Engineering Amrita School of Computing Bengaluru India

ISBN: (数字)9798331527549

ISBN: (纸本)9798331527556

Compiler bugs critically impact the correctness of software applications, making their detection and resolution vital for improving software quality. This paper proposes a novel approach leveraging sequential deep learning models, including Long short-term memory, Gated recurrent unit, Recurrent neural network, and a stacked ensemble model, to enhance compiler bug identification. By combining the strengths of these architectures, the proposed method improves predictive accuracy and generalization. To ensure model interpretability, we employ LIME (Local Interpretable Model-agnostic Explanations) to identify the features influencing bug detection decisions. Experimental results demonstrate that the stacked ensemble outperforms individual models in terms of precision, recall, Fl-score, and ROC-AUC. This work advances the state of compiler log analysis and contributes significantly to software quality assurance by integrating robust deep learning techniques with explainable AI.

关键词： Deep learning Analytical models Explainable AI computer bugs Software quality Feature extraction Data models Internet of Things Data communication Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Bangladeshi Road Traffic Sign Detection and Navigation Using Deep Learning

Bangladeshi Road Traffic Sign Detection and Navigation Using...

引用

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

作者： Tawhid Ahmed Komol Rinvi Jaman Riti Pulak Deb Nath Abdullah Al Noman Shayla Sharmin Sheak Rashed Haider Noori Department of Computer Science and Engineering Daffodil International University Dhaka Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Bangladeshi road traffic sign detection is vital for enhancing road safety and aiding autonomous driving systems by accurately identifying and interpreting local traffic signs. This research focuses on the development of deep learning techniques for Bangladeshi road traffic sign identification and Support, so as to improve the safety of roads and better organize traffic flow. The dataset 2710 raw images were selected and then augmented to 5420 images with specific traffic signs existing in Bangladesh containing 18 categories. The study aims to compare the results of four Transfer Learning model including Xception, VGG19, Inception-ResNetV2, and MobileNetV2 with a novel CNN architecture. Qualitative results reveal that Proposed CNN has achieved the maximum accuracy of 99.54%, surpassing other architectures. Most of the features developed to pre-process data such as normalization and data augmentation to enhance the quality of the dataset and improve the models performance.

关键词： Deep learning Computational modeling Transfer learning computer architecture Data augmentation Road traffic Data models Safety Convolutional neural networks Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Vision Transformer for Intracranial Hemorrhage Classification in CT Scans Using an Entropy-Aware Fuzzy Integral Strategy for Adaptive Scan-Level Decision Fusion

arXiv

引用

arXiv 2025年

作者： Chagahi, Mehdi Hosseini Delfan, Niloufar Moshiri, Behzad Piran, Md. Jalil Parikhan, Jaber Hatam School of Electrical and Computer Engineering College of Engineering University of Tehran Tehran Iran Department of Electrical and Computer Engineering University of Waterloo Waterloo Canada Department of Computer Science and Engineering Sejong University Seoul05006 Korea Republic of

Intracranial hemorrhage (ICH) is a critical medical emergency caused by the rupture of cerebral blood vessels, leading to internal bleeding within the skull. Accurate and timely classification of hemorrhage subtypes is essential for effective clinical decision-making. To address this challenge, we propose an advanced pyramid vision transformer (PVT)-based model, leveraging its hierarchical attention mechanisms to capture both local and global spatial dependencies in brain CT scans. Instead of processing all extracted features indiscriminately, A SHAP-based feature selection method is employed to identify the most discriminative components, which are then used as a latent feature space to train a boosting neural network, reducing computational complexity. We introduce an entropy-aware aggregation strategy along with a fuzzy integral operator to fuse information across multiple CT slices, ensuring a more comprehensive and reliable scan-level diagnosis by accounting for inter-slice dependencies. Experimental results show that our PVT-based framework significantly outperforms state-of-the-art deep learning architectures in terms of classification accuracy, precision, and robustness. By combining SHAP-driven feature selection, transformer-based modeling, and an entropy-aware fuzzy integral operator for decision fusion, our method offers a scalable and computationally efficient AI-driven solution for automated ICH subtype classification. © 2025, CC BY-NC-ND.

关键词： computerized tomography

来源：评论

学校读者我要写书评

暂无评论

Gen-C: Populating Virtual Worlds with Generative Crowds

arXiv

引用

arXiv 2025年

作者： Panayiotou, Andreas Charalambous, Panayiotis Karamouzas, Ioannis Department of Computer Science University of Cyprus Cyprus CYENS Centre of Excellence Cyprus Department of Computer Science and Engineering University of California Riverside United States

Over the past two decades, researchers have made significant advancements in simulating human crowds, yet these efforts largely focus on low-level tasks like collision avoidance and a narrow range of behaviors such as path following and flocking. However, creating compelling crowd scenes demands more than just functional movement—it requires capturing high-level interactions between agents, their environment, and each other over time. To address this issue, we introduce Gen-C, a generative model to automate the task of authoring high-level crowd behaviors. Gen-C bypasses the labor-intensive and challenging task of collecting and annotating real crowd video data by leveraging a large language model (LLM) to generate a limited set of crowd scenarios, which are subsequently expanded and generalized through simulations to construct time-expanded graphs that model the actions and interactions of virtual agents. Our method employs two Variational Graph Auto-Encoders guided by a condition prior network: one dedicated to learning a latent space for graph structures (agent interactions) and the other for node features (agent actions and navigation). This setup enables the flexible generation of dynamic crowd interactions. The trained model can be conditioned on natural language, empowering users to synthesize novel crowd behaviors from text descriptions. We demonstrate the effectiveness of our approach in two scenarios, a University Campus and a Train Station, showcasing its potential for populating diverse virtual environments with agents exhibiting varied and dynamic behaviors that reflect complex interactions and high-level decision-making patterns. © 2025, CC BY-NC-SA.

关键词： Digital elevation model

来源：评论

学校读者我要写书评

暂无评论

Internet of Things: An Aggregated Proof based Secure Healthcare Monitoring System- Comparative Analysis

Internet of Things: An Aggregated Proof based Secure Healthc...

引用

Mobile Computing and Sustainable Informatics (ICMCSI), International Conference on

作者： Aswathy R H Suresh P G. Jenifa Department of Computer Science and Engineering Saveetha School of Engineering SIMATS Thandalam Chennai India Computer Science and Engineering (Artificial Intelligence) Madanapalle Institute of Technology & Science Madanapalle Andhra Pradesh India

ISBN: (数字)9798331522667

ISBN: (纸本)9798331522674

The healthcare monitoring system plays a crucial role in remote monitoring. A highly secure healthcare system utilizing advanced cryptographic methods to protect sensitive information. The system integrates a multifactor authentication framework combined with aggregated proof mechanisms to enhance security. A detailed security analysis has been conducted to assess its effectiveness against various cyber threats. Compared to existing solutions, the proposed model demonstrates exceptional resilience to numerous known attacks, including Session Key Computation, Password Guessing, User & Gateway Node Impersonation, Replay Attacks, Sensor Node Spoofing, and Privileged Insider Attacks. These features ensure robust protection for healthcare data, addressing critical security concerns in modern applications. By leveraging the encryption and authentication strategies, the system provides a dependable and secure environment for the healthcare industry. This approach not only strengthens data privacy but also ensures operational integrity, making it a comprehensive solution for addressing current and emerging security challenges

关键词： Resistance Data privacy Computational modeling Authentication Medical services Passwords Logic gates Security Remote monitoring Resilience

来源：评论

学校读者我要写书评

暂无评论

A Blockchain-Assisted Attribute-Based Signcryption for Secure Sharing of Medical IoT Data in Fog Environment

引用

Security and Privacy 2025年第3期8卷

作者： Thushara, G.A. Bhanu, S. Mary Saira Department of Computer Science and Engineering National Institute of Technology Tiruchirappalli India

The growth of digital healthcare technologies has elevated the significance of sharing Medical Internet of Things (MIoT) data. By sharing information between healthcare providers, patients, and researchers, these technologies can enhance the accuracy of diagnoses, improve treatment effectively, and overall lead to better health care. Traditional cloud-based MIoT data-sharing methods are susceptible to security risks, latency issues, bandwidth consumption, and single points of failure. Moreover, traditional encryption methods in cloud-based systems are not feasible for MIoT data sharing due to complex key management and high computational requirements. Fog computing reduces latency by bringing data processing closer to the source, enabling decentralized computation. However, challenges persist in fog-based MIoT data sharing, including data access control, authentication, confidentiality, and unforgeability. To tackle these challenges, this paper proposes the integration of Ciphertext Policy Attribute-Based Signcryption (CP-ABSC) within a blockchain framework to facilitate fog-based MIoT data sharing. The proposed system divides the fog layer into two layers: the child fog layer focuses on secure access control using signcryption for confidential and unforgeable data sharing, while the parent fog layer deals with load balancing and Erasure Encoding Techniques (EET) to improve latency and storage efficiency. Using blockchain technology, unforgeable data-sharing mechanisms can provide assurances about the integrity and authenticity of shared data. In this paper, some security attacks are explored that compromise data confidentiality and unforgeability by altering its content. The proposed system's security deals with Elliptic Curve Computational Diffie-Hellman (ECCDH) and Elliptic Curve Decisional Diffie-Hellman (ECDDH) hard problems. Experimental analyses demonstrate its superiority in ensuring secure MIoT data sharing in resource-constrained applications. © 2025 John

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

A Novel Gain Modeling Technique for LLC Resonant Converters based on The Hybrid Deep-Learning/GMDH Neural Network

arXiv

引用

arXiv 2025年

作者： Mohammadi, Parham Electrical Engineering and Computer Science Department York University TorontoON Canada

This paper presents a novel hybrid approach for modeling the voltage gain ofLLC resonant converters by combining deep-learning neural networks with thepolynomial based Group Method of Data Handling (GMDH). While deep learningoffers high accuracy in predicting nonlinear converter behavior, it producescomplex network models. GMDH neural networks, in contrast, yield simpleralgebraic equations that can be more convenient in converter design. Bytraining a deep network on data from an FPGA based real time simulator and thenusing the network s predictions to train a GMDH model, the proposed hybridmethod achieves both high accuracy and design friendly simplicity. Experimentalresults show significant improvements over traditional methods such as FirstHarmonic Approximation (FHA) and frequency domain corrections, particularly forwide operating ranges. Copyright © 2025, The Authors. All rights reserved.

关键词： Resonant converters

来源：评论

学校读者我要写书评

暂无评论

Bridging the Emotional Gap in AI: A Study on Speech Emotion Recognition for Adaptive Human computer Interaction

Bridging the Emotional Gap in AI: A Study on Speech Emotion ...

引用

Artificial Intelligence and Data engineering (AIDE), International Conference on

作者： V S S L Deepak Janapa Sri Karthikeya Manjunadha Machiraju Bhargav Anjan Kalyan Yekula Lakshmi Raghunadh Karri Venkatesh Thanneru M Srinivas Department of Computer Science and Engineering Amrita Vishwa Vidyapeetham Amaravati India

ISBN: (数字)9798331527518

ISBN: (纸本)9798331527525

Speech Emotion Recognition (SER) is an advancement that has attracted a lot of interest because of its potential uses in intelligent systems, mental health monitoring, and human-computer interaction (HCI). Even with the progress made in AI-driven HCI, many systems are still unable to accurately sense and comprehend human emotions. Virtual assistants may carry out tasks based on spoken instructions, but they don't react well to user’s emotions, which results in less-than-ideal interactions. In order to close this gap, this study implements a speech emotion recognition algorithm that uses the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), and Toronto Emotional Speech Set (TESS) datasets to examine voice characteristics. The system makes use of cutting-edge deep learning methods including Long Short-Term Memory (LSTM) networks to capture temporal relationships in speech patterns and Convolutional Neural Networks (CNNs) for feature extraction from spectrograms which results in building a Hybrid Model. Traditional machine learning models may not be able to capture subtle emotional nuances in complex data, but they do offer faster processing times and easier implementations. Conversely, deep learning models, including 2D architecture such as CNNs and LSTMs, need more processing power but are better at handling large datasets and detecting subtle emotional cues. By leveraging these advancements, this research aims to enhance virtual assistants and similar systems to better recognize and respond to emotional cues in real time. Thus, Opening the door for a technological environment that is more user-centered and sympathetic.

关键词： Human computer interaction Emotion recognition Adaptation models Computational modeling Virtual assistants Speech recognition Mental health Convolutional neural networks Long short term memory Monitoring

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：