检索结果-内蒙古大学图书馆

International Conference on Artificial Intelligence in Information and communication (ICAIIC)

作者： Mohammad Rasoul Tanhatalab Manuel Forrer Michael Nelz Dep. of Information Engineering and Computer Science Trento University Italy ferrisol Co Graz Austria Nelo Intelligence Eichenau Germany

ISBN: (数字)9798331507022

ISBN: (纸本)9798331507039

Outlier removal is an important step in 3D point cloud data processing. Various methodologies for outlier removal will be analyzed shortly in this paper by describing a range of statistical and machine-learning techniques. Thus, the intent is to analyze different techniques and evaluate their performance in order to justify unique advantages and limitations of each technique. These limitations and advantages advance the discussion by introducing a new approach combining statistical and machine learning techniques on a real-world dataset. It not only identifies the spatial position of outliers but also enhances the measurement precision at those specific points. This work has a lot of significance from many aspects for researchers, engineers, and practitioners. Especially in robotics, industrial processes, augmented reality, and autonomous driving, the nature of 3D point cloud data requires sophisticated outlier handling. We believe our work will serve as a reference to optimize data quality in these dynamic applications under development.

关键词： Point cloud compression Three-dimensional displays Filtering Statistical analysis Service robots Machine learning Signal processing Position measurement Mathematical models Standards

来源：评论

学校读者我要写书评

暂无评论

An Implementation and Design of a Digital Light Switch of Smart-Home Based on MQTT Broker 4

An Implementation and Design of a Digital Light Switch of Sm...

引用

4th IEEE International Conference on Knowledge Innovation and Invention, ICKII 2021

作者： Tsao, Yung-Chung Wu, Chia-Chun Kuo, Yaw-Wen Tsai, Yin-Te Hsu, Chihcheng Wang, Renjie Chou, Polun Ho, Liulin Li, Chiyang Ko, Chihsheng National Chi Nan University Dep. of Electrical Engineering Puli Taiwan National Quemoy University Dep. of Industrial Engineering and Management Kinmen Taiwan Providence University Dep. of Computer Science and Communication Engineering Taichung Taiwan National Central Univeristy Dep. of Information Management Chung-Li Taiwan College of Science and Technology National Chi Nan University Taiwan

ISBN: (纸本)9781665423076

Home appliances are designed and manufactured for the market. Most users find that APPs are installed on smartphones or tablets to control home appliances with individual APP. Even light-bulbs with different brands or models need exclusive control software to corresponding light-bulbs. Thus, the smart and intelligent technology for smart-home is difficult to use due to different brands home-appliances from different manufactures. It lacks a standard to coordinate the control way like traditional light-bulbs-based E27 specs. For the light-bulbs of home appliances as an example, the traditional light-bulbs-based E27 lamp-holder have different fixtures and power-supply. Thus, new light-bulbs of home appliances for smart-home need a complex control-way besides electronic power to achieve smart functions and remote-control based on APPs on users' smart-phone. In this research, we try to integrate the internet of thing and message queue telemetry transport (MQTT) to design a mechanism and to implement a prototype web system to achieve an instant and visualized digital light switch of smart-home to solve the problems. For the validation of this research, we used a NodeMCU-32 microcontroller unit as a micro-controller to integrate with WS2812B LED strips to design light tube and light-bulb as the proof of concept (PoC) to demonstrate it in smart-home. © 2021 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

An Efficient Quantum cryptanalysis of lightweight hash function - PHOTON-Beetle 2

An Efficient Quantum cryptanalysis of lightweight hash funct...

引用

2nd International Conference on Automation, Computing and Renewable Systems, ICACRS 2023

作者： Allu, Swamy Naidu Tentu, Appala Naidu Kancharla, Gangadhara Rao Dep. of Computer Science & Engineering Acharya Nagarjuna University Andhra Pradesh Guntur522510 India CR Rao AIMSCS UoH Campus Telangana Hyderabad500046 India CR Rao Advanced Institute of Mathematics Statistics and Computer Science University of Hyderabad Campus Telangana Hyderabad500046 India

ISBN: (纸本)9798350340235

Quantum cryptanalysis methods are used to crack the majority of symmetric key ciphers. With this knowledge, cryptologists should be able to create post-quantum symmetric ciphers by incorporating these new quantum cryptanalysis techniques into their toolkit. Secure communication is crucial for applications related to the Internet of Things (IoT) to avoid cyberattacks. Cryptographic hash functions can be used to protect data integrity. The development of large-scale, fault-tolerant quantum computers is the most significant challenge. Recent techniques to attack symmetric key schemes can be achieved by using the Pre-image attack method, Simon’s algorithm, or by combining Simon’s and Grover’s algorithms. In this study, we explored some of the most recent methods for quantum cryptanalysis of symmetric key ciphers. Firstly, we discussed the design of quantum cryptanalysis methods, along with the advantages over classical procedures. In this paper, we illustrate the implementation of the NIST-standard lightweight hash function PHOTON-Beetle on a Quantum computer (using IBM Qiskit), which is mostly used in IoT devices. Efficient quantum implementation of PHOTON-Beetle is proposed in this paper. This paper calculates the quantum cost of PHOTON-Beetle for every layer and compares it with [1]. © 2023 IEEE.

关键词： IoT PHOTON-Beetle Post-quantum cryptography Quantum algorithms Quantum cryptanalysis Symmetric cryptanalysis

来源：评论

学校读者我要写书评

暂无评论

Biological age estimation from ECG using Artificial Intelligence approaches

Biological age estimation from ECG using Artificial Intellig...

引用

Biomedical engineering Conference (JIBEC), Jordanian International

作者： Jana Sleiman Mohamad Daher Mohamad Khalil Ahmad Diab Abdallah Kassem Computer and Communications Engineering Lebanese University Tripoli Lebanon Dep. of Computer Science/Dep. of Computer Science & Information Technology Faculty of Science/Faculty of Applied Sciences & Technology Tripoli Lebanon Electrical Engineering Lebanese University Tripoli Lebanon Biomedical Engineering Lebanese International University Beirut Lebanon ECCE Dept. Faculty Of Engineering Notre Dame University Zouk Mosbeh Lebanon

ISBN: (数字)9798331539542

ISBN: (纸本)9798331539559

This paper explores the application of Artificial Intelligence (AI) techniques in healthcare, specifically focusing on electrocardiogram (ECG) data analysis and biological age estimation. The study begins with an overview of ECG as a diagnostic tool and discusses the integration of Deep Learning (DL) models, particularly Convolutional Neural Networks (CNNs), in healthcare for various applications. Previous studies reviewed in this context consistently identify CNNs as the superior model for ECG analysis and prediction tasks. The primary research investigates the use of Machine Learning (ML) algorithms to estimate biological age from ECG data, employing metrics such as Mean Absolute Error (MAE) and variance to evaluate model performance. Subsequently, DL models are implemented and compared, demonstrating superior accuracy in estimating the biological age of new patients. While both ML and DL approaches were initially explored, this paper primarily focuses on deep learning methodologies due to their demonstrated superiority in healthcare applications. The findings highlight the efficacy of deep learning over traditional ML methods, underscoring its potential for enhancing diagnostic accuracy and personalized patient care. For instance, the most accurate deep learning model, the Convolutional Neural Network (CNN), achieved a Mean Absolute Error (MAE) of 1.49 and an R 2 score of 0.53, whereas the best-performing traditional ML model, Random Forest, achieved an MAE of 1.54 and an R 2 score of 0.55. These results clearly indicate the advantage of deep learning models in capturing complex patterns in ECG data, leading to more precise and reliable estimations of biological age.

关键词： Deep learning Maximum likelihood estimation Accuracy Data analysis Biological system modeling Medical services Electrocardiography Data models Convolutional neural networks Random forests

来源：评论

学校读者我要写书评

暂无评论

Machine learning algorithms in Bigdata Analysis and its applications: A review 4

Machine learning algorithms in Bigdata Analysis and its appl...

引用

4th International Congress on Human-computer Interaction, Optimization and Robotic Applications, HORA 2022

作者： Khoshaba, Farah Kareem, Shahab Awla, Hoshang Mohammed, Chnar Erbil Polytechnic University Technical Engineering College Information System Engineering Department Erbil Iraq Lebanese French University College of Engineering and Computer Science Dep. of Information Technology Kurdistan Erbil Iraq Soran University College of Science Department of Computer Science Erbil Iraq

ISBN: (纸本)9781665468350

A wide range of disparate variety of heterogeneous and even disparate data sources has been integrated into the computer science research principles through the assistance of Artificial Intelligence (AI) and Machine Learning (ML) delivering excellent results in accuracy and data quality. Even so, taking machine learning approaches to very broad and dynamic datasets is both logistically and computationally costly. Some people are generating so much data regularly, which means that advanced analytics platforms have greater importance than ever before. Spark Machine Learning has extensive capabilities for a variety of ML activities such as regression, grouping, dimension reduction, and extraction of rules. While it has numerous datasets of high quality and consistency of structure. The specification of the MapReduce programming model was intended for the massive density of large-volume computation. It accomplishes this by splitting the job into several well-defined sub-sized units. There are various sets of data mining frameworks that link to a distributed device. Hadoop serves as the infrastructure for the distributed system. Several resources claim that Mahout will scale to handle massive datasets. This article discusses and demonstrates the use of algorithms and their significant impact on Bigdata, as well as a comparison between their implementations among Apache Spark and Apache Mahout. © 2022 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Secure Password Using EEG-based BrainPrint System: Unlock Smartphone Password Using Brain-computer Interface Technology

Secure Password Using EEG-based BrainPrint System: Unlock Sm...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Alkhyeli, Zuwaina Alshehhi, Ayesha Alhemeiri, Mazna Aldhanhani, Salma Albalushi, Khalil Alnuaimi, Fatima Ali Belkacem, Abdelkader Nasreddine UAE University CIT Dep. Computer Science and Software Engineering Al Ain United Arab Emirates UAE University CIT Dep. Information Systems and Security Al Ain United Arab Emirates United Arab Emirates University CIT Dep. Computer & Network Engineering Al Ain United Arab Emirates Autonomous Robotics Research Center TII Abu Dhabi United Arab Emirates

ISBN: (纸本)9781665468190

As security becomes a strong factor in daily activities, finding secure ways to unlock machines and smartphones is a challenge due to hardware limitations and the high risk of hacking. Considering the level of security and privacy in the digital world, attackers tend to be one step ahead. Therefore, this technical paper introduces a brain-computer interface (BCI) for increasing subject-based security using unique biometric features as a solution to build complex passwords. The BCI measures brain changes and extracts relevant bio-features from each subject using non-invasive electroencephalogram (EEG) tests. The proposed system allows users to gain access to their devices using brain waves (bypass) instead of inserting their password manually (normal path), which saves the user time and upgrades the level of privacy as no physical actions are required during this process. This system is also well suited for individuals with mobility impairments. We used the P300-based BCI controlling paradigm which dep.nds on reading the electrical brain activity of the user when observing a particular object. The other feature of the system is that it can extract unique features of each individual brain to produce a network that uniquely identifies them, which is used as a security layer. Users need to enter their unique network to access their device with failed attempts requiring an EEG test to identify the user. The system plays an active role in facilitating user processes for authentication while accessing devices. The system establishes an urgent call whenever the user's brain currents command it to. The project outcomes were assessed by simulating the BCI before real-time implementation to determine errors and resolve the validity of the project scope. © 2022 IEEE.

关键词： Brain computer interface

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Impact of Injected Mobility Data on Measuring Data Coverage in CrowdSensing Scenarios

Evaluating the Impact of Injected Mobility Data on Measuring...

引用

IEEE Conference on Global communications (GLOBECOM)

作者： Alexander Kocian Michele Girolami Stefano Capoccia Luca Foschini Stefano Chessa Dep. of Computer Science University of Pisa Italian National Council of Research ISTI-CNR Pisa Italy Department of Computer Science and Engineering University of Bologna Bologna Italy

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

A major weakness of Mobile CrowdSensing Platforms (MCS) is the willingness of users to participate, as this implies disclosing their private data (for example, concerning mobility) to the MCS platform. In the effort to enforce data privacy in the creation of mobility coverage maps using an MCS platform, recent work proposes the use of a spatially distributed approach that, however, is vulnerable to data injection attacks. In this contribution, we define and implement a progressive attacker model following a statistical approach. We propose a novel mitigation strategy based on unsupervised anomaly detection. Accessing the coverage performance with real-world mobility data indicates that the mean value of the attacker’s profile determines the probability of being revealed. In particular, we are able to identify the attacker and filter out the data injected by the attackers with high precision.

关键词： Data privacy Crowdsensing Prevention and mitigation Distributed databases Probability Data models Security Proposals Mobile computing Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Maximize Leader Election Algorithm Performance in Bi-Directional Ring Networks With Both Links Down Between Two Nodes

Maximize Leader Election Algorithm Performance in Bi-Directi...

引用

Information Technology (ACIT)

作者： Mohammed N. Al Refai Hamed Fawareh Hamzeh Aljawawdeh Fuad Sameh Alshraiedeh Mohammad Al-quraan Hisham Al-Odat Ahmed Ali Otoom Zeyad M. Jamhawi Mohammad Kanan Software Engineering Dep. Zarqa University Zarqa Jordan Computer Science Zarqa University Zarqa Jordan Department of Cyber Security Applied Science Private University Amman Jordan Department of Computer Science Applied Science Private University Amman Jordan Dep. of Industry Engineering University of Business and Technelogy Jeddah Saudi Arabia

ISBN: (数字)9798331540012

ISBN: (纸本)9798331540029

The process of selecting a leader to avoid instability in distributed system network after leader failure is called Leader Election Algorithm (LEA). This process becomes more complicated when it faces intermittent or complete links disconnected. the network becomes a one-way behavior if both lines of communication between two points are disrupted and thus the network becomes very slow, especially in systems where time is a critical issue, and the process of communication between points becomes much slower if the network leader is next to the fault or close to it, so the criteria will be established to ensures that the system leader in the Bidirectional ring communication network is in a place that allows next leader to communicate with all points as quickly as possible, and this will be during the election *** research presents a solution for leader failure problem in bidirectional ring network. This research takes into account the existence of intermittent or disconnection links between nodes. It will produce three algorithms to determine a new leader in the network with maximum network performance. It uses the least probability of time steps and the number of messages needed to elect the new leader in bidirectional ring network with N communicated nodes. so that the system remains able to work and communicate with its best capacity during this problem with the number of message complexity O(N) in the worst case. The third scenario minimizes the value of proper leader ID $\frac{{I{D_{Current}}}}{{{I_{Current}}idStepCount}}$, and maximizes the value of proper leader place N − (N − nStepCount).

关键词： Voting Bidirectional control Complexity theory communication networks Information technology Faces

来源：评论

学校读者我要写书评

暂无评论

RF-Based Lightweight Machine Learning for Comprehensive Drone Activity Classification

RF-Based Lightweight Machine Learning for Comprehensive Dron...

引用

Artificial Intelligence, Blockchain, and Internet of Things (AIBThings), IEEE International Conference on

作者： Mohammed A. Alqodah Mahdia Tahsin Mohammad H. Omari Mustafa M. Matalgah David Harrison Dep. of Electrical & Computer Engineering University of Mississippi University MS USA Dept. of Computer and Information Science University of Mississippi University MS USA

ISBN: (数字)9798331529529

ISBN: (纸本)9798331529536

The rapid rise of commercially available drones, driven by miniaturization and affordability, has introduced potential security risks alongside numerous societal benefits. While drones serve various industries, their unregulated use necessitates robust detection and classification methods. This paper proposes a novel, efficient model for classifying unwanted drone activity. We leverage statistical and time-domain features extracted from radio frequency (RF) signals and machine learning (ML) classifiers. This approach aims to not only detect drones but also distinguish their models and operational modes. Utilizing the DroneRF dataset, our feature selection strategy prioritizes informative signals in the low-band spectrum, achieving ${100\%}$ accuracy in both drone detection and drone model identification tasks. Furthermore, our hybrid model, combining time-domain and frequency-domain features, achieves an accuracy of ${97.66\%}$ in classifying different flight modes. This demonstrates the effectiveness of this comprehensive feature extraction approach for drone activity classification.

关键词： Radio frequency Accuracy RF signals Neural networks Feature extraction Real-time systems Security Time-domain analysis Random forests Drones

来源：评论

学校读者我要写书评

暂无评论

CLIP-VAD: Exploiting Vision-Language Models for Voice Activity Detection

arXiv

引用

arXiv 2024年

作者： Appiani, Andrea Beyan, Cigdem Dep. of Management Information and Production Engineering University of Bergamo Bergamo Italy Department of Computer Science University of Verona Verona Italy

Voice Activity Detection (VAD) is the process of automatically determining whether a person is speaking and identifying the timing of their speech in an audiovisual data. Traditionally, this task has been tackled by processing either audio signals or visual data, or by combining both modalities through fusion or joint learning. In our study, drawing inspiration from recent advancements in visual-language models, we introduce a novel approach leveraging Contrastive Language-Image Pretraining (CLIP) models. The CLIP visual encoder analyzes video segments composed of the upper body of an individual, while the text encoder handles textual descriptions automatically generated through prompt engineering. Subsequently, embeddings from these encoders are fused through a deep neural network to perform VAD. Our experimental analysis across three VAD benchmarks showcases the superior performance of our method compared to existing visual VAD approaches. Notably, our approach outperforms several audio-visual methods despite its simplicity, and without requiring pre-training on extensive audio-visual datasets. © 2024, CC BY-NC-ND.

关键词： Visual languages

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：