检索结果-内蒙古大学图书馆

27th International Conference on Pattern Recognition, ICPR 2024

作者： Siddhad, Gourav Dey, Sayantan Roy, Partha Pratim Iwamura, Masakazu Department of Computer Science and Engineering Indian Institute of Technology Roorkee Uttarakhand Roorkee247667 India Department of Core Informatics Graduate School of Informatics Osaka Metropolitan University Osaka Sakai599-8531 Japan

ISBN: (纸本)9783031781940

Driver fatigue detection is increasingly recognized as critical for enhancing road safety. This study introduces a method for detecting driver fatigue using the SEED-VIG dataset, a well-established benchmark in EEG-based vigilance analysis. By employing advanced pattern recognition technologies, including machine learning and deep neural networks, EEG signals are meticulously analyzed to discern patterns indicative of fatigue. This methodology combines feature extraction with a classification framework to improve the accuracy of fatigue detection. The proposed NLMDA-Net reached an impressive accuracy of 83.71% in detecting fatigue from EEG signals by incorporating two novel attention modules designed specifically for EEG signals, the channel and depth attention modules. NLMDA-Net effectively integrate features from multiple dimensions, resulting in improved classification performance. This success stems from integrating temporal convolutions and attention mechanisms, which effectively interpret EEG data. Designed to capture both temporal and spatial characteristics of EEG signals, deep learning classifiers have proven superior to traditional methods. The results of this study reveal a substantial enhancement in detection rates over existing models, highlighting the efficacy of the proposed approach for practical applications. The implications of this research are profound, extending beyond academic realms to inform the development of more sophisticated driver assistance systems. Incorporating this fatigue detection algorithm into these systems could significantly reduce fatigue-related incidents on the road, thus fostering safer driving conditions. This paper provides an exhaustive analysis of the dataset, methods employed, results obtained, and the potential real-world applications of the findings, aiming to contribute significantly to advancements in automotive safety. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimal Resource Allocation in Fog Computing for Healthcare Applications

引用

computers, Materials & Continua 2022年第6期71卷 6147-6163页

作者： Salman Khan Ibrar Ali Shah Nasser Tairan Habib Shah Muhammad Faisal Nadeem Department of Computer Software Engineering University of Engineering and TechnologyMardan23200Pakistan Department of Computer Science College of Computer ScienceKing Khalid UniversityAbhaK.S.A Informatics Complex H-8Islamabad44000Pakistan

In recent years,the significant growth in the Internet of Things(IoT)technology has brought a lot of attention to information and communication *** IoT paradigms like the Internet of Vehicle Things(IoVT)and the Internet of Health Things(IoHT)create massive volumes of data every day which consume a lot of bandwidth and ***,to process such large volumes of data,the existing cloud computing platforms offer limited resources due to their distance from IoT ***,cloudcomputing systems produce intolerable latency problems for latency-sensitive real-time ***,a newparadigm called fog computingmakes use of computing nodes in the form of mobile devices,which utilize and process the real-time IoT devices data in orders of *** paper proposes workload-aware efficient resource allocation and load balancing in the fog-computing environment for the *** proposed algorithmic framework consists of the following components:task sequencing,dynamic resource allocation,and load *** consider electrocardiography(ECG)sensors for patient’s critical tasks to achieve maximum load balancing among fog nodes and to measure the performance of end-to-end delay,energy,network consumption and average *** proposed algorithm has been evaluated using the iFogSim tool,and results with the existing approach have been *** experimental results exhibit that the proposed technique achieves a 45%decrease in delay,37%reduction in energy consumption,and 25%decrease in network bandwidth consumption compared to the existing studies.

关键词： Fog computing internet of health things resource management

来源：评论

学校读者我要写书评

暂无评论

Enhancing Sign Language Recognition Using Deep Convolutional Neural Networks 14

Enhancing Sign Language Recognition Using Deep Convolutional...

引用

14th International Conference on Information, Intelligence, Systems and Applications, IISA 2023

作者： Kanavos, Athanasios Papadimitriou, Orestis Mylonas, Phivos Maragoudakis, Manolis University of the Aegean Department of Information and Communication Systems Engineering Samos Greece University of West Attica Department of Informatics and Computer Engineering Athens Greece Ionian University Department of Informatics Corfu Greece

ISBN: (纸本)9798350318067

The advancements in sensing technologies and AI algorithms have opened up a wide range of possibilities for developing applications to meet the needs of individuals who are deaf or hard of hearing. Sign language plays a vital role in the lives of people with hearing and speaking disabilities. This research aims to explore digital image processing and machine learning methods for efficiently building a sign language dataset and creating a sign language interface system. The proposed system utilizes a Convolutional Neural Network (CNN) to analyze and interpret hand gestures and poses, converting them into natural language. The developed CNN model specifically focuses on improving the accuracy of predicting the American Sign Language alphabet. Despite variations in dataset conditions and size, the model achieved an exceptional accuracy rate of 98.73%. Additionally, it demonstrated a low loss value of 0.0539, indicating its robust performance. © 2023 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Disease Classification in the Cotton Using Convolution Neural Networks

Disease Classification in the Cotton Using Convolution Neura...

引用

International Conference on Advances in Data Science and Computing Technologies, ASDC 2022

作者： Sharma, Aviral Chhabra, Gunjan School of Computer Science Department of Informatics UPES Uttarakhand Dehradun India Department of Computer Science and Engineering GEHU Uttarakhand Dehradun India

ISBN: (纸本)9789819936557

The present research focuses on an integrated architecture for extraction and identification of disease in the cotton plants. The deep learning architecture provided in the paper is capable of extracting the features from the cotton images and then create feature map of the same. The methods provide an automated approach for disease detection using image parts of various plant parts. The proposed methodology provides a unified approach for detection of major cotton diseases, namely bacterial blight, cotton leaf curl virus and wilt. The model can implemented in mobile applications for mass deployments and timely intervention be provided for protection of crops. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Cotton

来源：评论

学校读者我要写书评

暂无评论

Jointly Preserving Subject and Object Anonymity in Nearby Friends Querying for Curious Service Providers

Jointly Preserving Subject and Object Anonymity in Nearby Fr...

引用

2024 Cognitive Models and Artificial Intelligence Conference, AICCONF 2024

作者： Abul, Osman Department of Computer Science College of Computing and Informatics University of Sharjah Sharjah United Arab Emirates Department of Computer Engineering TOBB University of Economics and Technology Ankara Ankara Turkey

ISBN: (纸本)9798400716928

This work studies the urban area location privacy preserving location updates and nearby friends (NF) querying for the centralized proximity based services (PBSs). The urban area constraint forces the user mobility to be on the city road network, modeled as a weighted directed graph. The location privacy model exploits location anonymity through subgraph cloaking. Two kinds of location privacy issues have been identified and handled in a unified framework, (i) Subject Anonymity on location updates, i.e., providing location anonymity of the location updater (against the curious PBS provider), and (ii) Object Anonymity on NF querying, i.e., providing location anonymity of friends in NF query responses (against the curious query issuer). In the former, the user simulates the PBS provider's belief (a subgraph) on his location whereabouts, and locally blocks anonymity violating location update requests which is understood by a simulated attack. In the latter, as the NF query processor, the PBS provider simulates the user's mutual beliefs (subgraphs) on friends' location whereabouts obtained through previous NF queries. By a simulated attack, some friends' existence are suppressed from the NF query responses. An experimental evaluation on a real dataset has been conducted. © 2024 ACM.

关键词： Location

来源：评论

学校读者我要写书评

暂无评论

Lightweight image super-resolution reconstruction based on mixed attention and global inductive bias network

引用

Multimedia Tools and Applications 2024年 1-17页

作者： Cai, Yuxi Wen, Xiaopeng Wang, Tongguan College of Computer Science Sichuan University Chengdu China Department of Electronic Information and Electrical Engineering Dalian University of Technology Dalian China College of Informatics Huazhong Agricultural University Wuhan China

Convolutional neural network has significantly advanced the field of image super-resolution reconstruction in recent years. The insufficient ability to model global information of hierarchical features, incomplete attention to information, and excessive parameters make some based on convolutional methods face challenges. To address these issues, this paper designs the lightweight image super-resolution reconstruction based on Mixed Attention and Global Inductive bias Network(MAGIN), which directly captures the global information of phased features from the horizontal and vertical directions respectively to produce global inductive bias information. Furthermore, the global spatial attention block is proposed, which can realize the perception of global feature space without redundant operation. To realize more comprehensive attention for features, a mixed attention block arises, which enables the network to attend to features across different dimensions: channel, spatial, and pixel levels. Extensive experimental results demonstrate that the MAGIN delivers more satisfactory results than some other advanced lightweight reconstruction algorithms on several standard test datasets under the condition of considering the network’s performance, parameter and computational complexity. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

GestureSet: Public Domain Dataset for Human Gesture Recognition using Wrist-worn Devices: A Preliminary Version 19

GestureSet: Public Domain Dataset for Human Gesture Recognit...

引用

19th Annual International Conference on Distributed Computing in Smart Systems and the Internet of Things, DCOSS-IoT 2023

作者： Tzamalis, Pantelis Nikoletseas, Sotiris Spirakis, Paul G. University of Patras Computer Engineering and Informatics Department Patras Greece Patras Greece University of Liverpool Department of Computer Science United Kingdom

ISBN: (纸本)9798350346497

Human Gesture Recognition (HGR) has become an increasingly important research area in recent years, driven by the need for more natural and intuitive ways of interacting with machines. Wearable smart devices, such as smartwatches, smart wristbands, and other wrist-worn devices have the potential to enable accurate and efficient gesture recognition in real-world settings. However, equivalent problems, such as object detection in computer vision and audio identification in audio analysis, have obtained enormous advantages from comprehensive datasets over the last years, for instance, ImageNet and AudioSet, accordingly. This paper proposes the structure and the creation of a preliminary version of GestureSet, an HGR large-scale multitask dataset, which includes manually annotated data of gesture events that were collected from wrist-worn devices. In particular, it proposes a hierarchy abstraction schema for the proper dataset structure and instantiates the basis of the dataset through the first HGR task that is related to the collection and annotation of gestures related to allergic rhinitis identification. Thus, through this preliminary version, we describe how the proper dataset structure is divided into different groups, each one including sessions, individual gestures, and pilot data that could be used for evaluating the learning algorithms' performance in real-world scenarios. To the best of our knowledge, this is the first attempt that envisions a unified dataset for various HGR tasks that can find applications in different domains. Our vision, specifically, is for this work to set the basis for the dataset enrichment in a wide range of various tasks, and its data to be used in multiple learning and identification settings, providing circularity in different domains, and enabling reproducibility of the dataset's gestures events. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Classification of Alzheimer’s Disease Subjects from MRI Using Deep Convolutional Neural Networks 3rd

Classification of Alzheimer’s Disease Subjects from MRI Usi...

引用

3rd International Conference on Novel and Intelligent Digital Systems, NiDS 2023

作者： Papadimitriou, Orestis Kanavos, Athanasios Mylonas, Phivos Maragoudakis, Manolis Department of Information and Communication Systems Engineering University of the Aegean Samos Greece Department of Informatics and Computer Engineering University of West Attica Athens Greece Department of Informatics Ionian University Corfu Greece

ISBN: (纸本)9783031441455

The classification of Alzheimer’s disease (AD) using deep learning techniques has shown promising results. However, achieving successful application in medical settings requires a combination of high precision, short processing time, and generalizability to different populations. In this study, we propose a convolutional semantic network (CNN)-based classification algorithm that utilizes magnetic resonance imaging (MRI) scans from individuals with AD. Our models achieved average area under the curve (AUC) values of 0.91−0.94 for within- dataset recognition and 0.88−0.89 for between-dataset recognition. The proposed convolutional framework can be potentially applied to any image dataset, offering the flexibility to design a computer-aided diagnosis system targeting the prediction of various clinical conditions and neuropsychiatric disorders using multimodal imaging and tabular clinical data. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Accurate Deep Learning-Based Sleep Apnea Detection from Cardiac Physiological Signals 8

Accurate Deep Learning-Based Sleep Apnea Detection from Card...

引用

8th IEEE Asia Pacific Conference on Wireless and Mobile, APWiMob 2023

作者： Delimayanti, Mera Kartika Muharram, Asep T. Pradipta, Ayres Ismail, Muhammad K. Ryanari, Rinaldito A. Prasetyo, Raditya A. Faisal, Mohammad Reza Politeknik Negeri Jakarta Department of Computer and Informatics Engineering Depok Indonesia Lambung Mangkurat University Department of Computer Science Banjarbaru Indonesia

ISBN: (纸本)9798350327861

Sleep apnea is a severe sleep disorder characterized by interrupted breathing during sleep. Early detection and accurate diagnosis of sleep apnea are essential to avoiding potentially dangerous health complications. This study proposes a sleep apnea detection model using cardiac physiological signals and a deep learning approach. This model uses Convolutional Neural Networks (CNN) to learn patterns and features in cardiac signal data. Heart signal data is collected from individuals with sleep apnea and normal sleeping individuals as datasets. Datasets consisting of heart signal data were gathered from patients diagnosed with sleep apnea and people with normal sleeping patterns. The public dataset was enhanced and utilized for training, employing a Convolutional Neural Network (CNN) architecture of six layers. The improved data is derived from the process of segmenting the data. The trained model demonstrated a high accuracy of 98.4% in effectively classifying physiological heart signal samples into two categories: 'normal' and 'sleep apnea'. This project aims to develop a model that can assist medical professionals in diagnosing sleep apnea and providing early warning to patients. © 2023 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Advancing Weather Image Classification Using Deep Convolutional Neural Networks 18

Advancing Weather Image Classification Using Deep Convolutio...

引用

18th International Workshop on Semantic and Social Media Adaptation and Personalization, SMAP 2023

作者： Papadimitriou, Orestis Kanavos, Athanasios Mylonas, Phivos Maragoudakis, Manolis University of the Aegean Department of Information and Communication Systems Engineering Samos Greece University of West Attica Department of Informatics and Computer Engineering Athens Greece Ionian University Department of Informatics Corfu Greece

ISBN: (纸本)9798350327717

Automatic classification of weather images is a crucial task in the field of meteorology. Recent advancements in Convolutional Neural Networks (CNNs) have demonstrated their effectiveness in image classification. In this paper, we propose an innovative CNN-based approach for weather image classification. Building upon prior research, our model capitalizes on the exceptional feature extraction capabilities of CNNs to accurately classify weather images across various categories. To enhance the generalization performance, we train the model using a combination of data augmentation techniques, including rotation, scaling and flipping. This paper introduces a CNN model specifically made for weather image classification, encompassing categories such as cloudy, sunny, rainy and snowy conditions. Through extensive training and evaluation on a substantial weather image dataset, our proposed model achieves an impressive accuracy rate of 98%. Our experimental results highlight the superior performance of our CNN model when compared to various state-of-The-Art methods in weather image classification. © 2023 IEEE.

关键词： Image classification

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：