检索结果-内蒙古大学图书馆

6th International Conference on Robotics and Automation in Industry, ICRAI 2024

作者： Qaiser, Muhid Abbas, Muhammad Sohail Muhammad, Asif Haider, Khadija Khan, Haroon National University of Computer and Emerging Sciences Department of Artificial Intelligence and Data Science Islamabad Pakistan National University of Computer and Emerging Sciences Department of Software Engineering Islamabad Pakistan

ISBN: (纸本)9798350355703

Retinal diseases such as Choroidal Neovascularization (CNV), Drusen, and Diabetic Macular Edema (DME) can severely impair vision and are challenging to reverse if detected late. Optical Coherence Tomography (OCT) serves as a non-invasive imaging modality to diagnose these conditions through cross-sectional retinal images. Current advancements in retinal disease detection predominantly focus on fine-tuning larger models to achieve high accuracy, often at the expense of substantial training time and data requirements. We introduce an innovative Convolutional Neural Network (CNN) named TezNet that is deeper and wider than existing custom CNNs for the dataset used in this research yet maintains a lightweight architecture with only 3.6 million parameters. This high efficiency is accomplished by leveraging depthwise and spatial separable convolutional techniques without employing preprocessing techniques or noise reduction filters. The data augmentation techniques generate diverse image qualities, ensuring robust predictions across various scenarios. By training exclusively on a single dataset, our model achieves rapid training times and optimal resource utilization, minimizing the risk of overfitting. This blend of advanced parameter optimization, minimal preprocessing, and strategic data augmentation underscores the potential of our CNN model for swift, precise, and resource-efficient retinal disease detection. TezNet model achieves a remarkable 96% accuracy, competing closely with state-of-the-art networks like ResNet50 (over 23M parameters), InceptionV3 (over 25M parameters), and ResNeXt (89M parameters). © 2024 IEEE.

关键词： Optical coherence tomography

来源：评论

学校读者我要写书评

暂无评论

A computer Vision-Based Vehicle Detection System Leveraging Deep Learning 1

A Computer Vision-Based Vehicle Detection System Leveraging ...

引用

1st IEEE Karachi Section Humanitarian Technology Conference, Khi-HTC 2024

作者： Khan, Saadullah Vohra, Sanayna Siddique, Sultan Abdullah Abro, Abdul Ahad Ebrahim, Mansoor Iqra University Main Campus Department of Software Engineering Karachi Pakistan Iqra University Main Campus Department of Computer Science Karachi Pakistan

ISBN: (纸本)9798350373042

This study uses advanced computer vision (CV) and deep learning (DL) techniques to present a combined approach for extensive vehicle detection and number plate recognition. Improving the effectiveness and precision of traffic surveillance and administration systems is the main goal. The two primary parts of the suggested system are identification of number plates recognition and vehicle detection. The vehicle detection module uses rigid images or video streams to recognize along with monitor vehicles in real-time using innovative DL algorithms. Such limitations of system reliability are overcome through the application of CNN and object detection models in an integrated computer vision system that allows for reliable performance under a variety of environmental conditions including unfavorable lighting and weather situations. The goal of an identification plate recognition feature is to read and extract alphanumeric characters from license plates that have been detected. To precisely identify and interpret license plate data, this stage combines DL models with the use of optical character recognition (OCR) techniques. The system can adapt to different geographical locations by supporting changes in plate layouts, fonts, and positions. By combining these two modules, a complete intelligent transportation system solution is produced, offering useful information for applications such as congestion examination, safety, and assurance. Because of its modular architecture, the system is easily scalable and customizable, enabling it to be adjusted to various hardware configurations and surveillance scenarios. Using real-world scenarios and benchmark datasets, extensive experiments are conducted to assess the effectiveness of the proposed system. The outcomes show excellent efficiency and accuracy in tasks involving the detection of vehicles and number plate recognition. By promoting enhanced security, integrity, and general traffic management, the suggested integrated syste

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Decentralized Face Identification with Hierarchical Navigable Small World on Blockchain

Decentralized Face Identification with Hierarchical Navigabl...

引用

2023 International Conference on Machine Learning and Automation, CONF-MLA 2023

作者： Lee, Hsiang-Hung Chen, Yiting Department of Computer Science & Information Engineering National Taiwan Normal University Taipei City106 Taiwan Department of Computer Software and Theory Donghua University Shanghai201620 China

This paper presents a novel method for decentralized storage in deep-learning-based face recognition systems using the Hierarchical Navigable Small World (HNSW) algorithm. The proposed solution utilizes Ethereum smart contracts, which acts as highly available data storage systems for storing identifiable data for authorized personnel. In addition, the solution is integrated with a centralized vector database that is in charge of vector indexing, searching and associating face embeddings to an identity on the Ethereum blockchain with anonymous hashes. Vector indexing and search processes involve different machine learning algorithms that enable computations to be carried out in a reasonable time with good matching accuracy. Specifically, we compared different approaches and selected the HNSW algorithm. Accordingly, we successfully implemented a prototype of a reliable and privacy-focused decentralized face identification system for areas under government surveillance, such as customs inspection sites. In our measurements, the system could handle 20,000 face vectors easily with high matching accuracy, and the performance could be further improved using more powerful hardware. Finally, we also propose additional methods to further scale up the system to handle millions of face vectors. © 2024 Institute of Physics Publishing. All rights reserved.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Internet of Things (IoT) applications security trends and challenges

引用

Discover Internet of Things 2024年第1期4卷 1-22页

作者： Laghari, Asif Ali Li, Hang Khan, Abdullah Ayub Shoulin, Yin Karim, Shahid Khani, Muhammad Adnan Kaim Software College Shenyang Normal University Shenyang China Department of Computer Science Bahria University Karachi Campus Karachi75000 Pakistan School of Information Engineering Xi’an Eurasia University Shaanxi Xian710065 China Department of Computer Science Ilma University Karachi Pakistan

The idea behind layered design is the foundation of the Internet of Things. Each tier uses a variety of technologies for capacity, preparation, and information transmission. With regard to the risks and vulnerabilities related to IoT-enabled devices, as well as potential assurance mechanisms in light of equipment constraints and novel information transfer techniques, this study attempts to analyze the current architecture of the Internet of Things. After that, we talk about IoT architecture and applications. The following is a list of effective real-time IoT applications that are now in use: Self-driving cars, traffic management systems, smart grids, logistic management hierarchies, environment monitoring, building safety applications, and many more are examples of emerging technologies. Specifically, we highlight the latest improvements and challenges associated with IoT security. Finally, we discuss the state-of-the-art research and its conclusions, as well as the outstanding problems for future IoT security research. © The Author(s) 2024.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

User Recognition based on Gait Pattern via Smartwatch Accelerometer in Unrestricted Environment 3

User Recognition based on Gait Pattern via Smartwatch Accele...

引用

3rd International Conference on Communication, Computing and Digital Systems, C-CODE 2023

作者： Khan, Sheharyar Shah, Syed M Adnan Arsalan, Aamir Ahmed, Wakeel University of Engineering and Technology Taxila Department of Computer Science Taxila Pakistan Fatima Jinnah Women University Rawalpindi Department of Software Engineering Rawalpindi Pakistan

ISBN: (纸本)9798350332391

Wearable gadgets are becoming a major constituent of our society due to a wide range of applications like financial transactions, unlocking automobiles, tracking health and fitness, and many more. Personal data is usually required to manage these services, however, most commercially available wearables devices either lack a user authentication mechanism or employ knowledge-based authentication like passwords, PINs, or pattern locks. In this growing digital age, it is really difficult for an individual to memorize such a huge number of passwords. In this research, we put forward a technique for recognizing users from their walking patterns recorded in real-life settings using a smartwatch accelerometer sensor. Ensemble classification techniques (bagging and boosting) are used for the purpose of user identification using the selected time domain features. Our proposed method attained the highest classification accuracy of 88.35% which transcends the existing user recognition schemes available in the literature. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection for Wireless Sensor Network Using Graph Neural Networks

Intrusion Detection for Wireless Sensor Network Using Graph ...

引用

2023 IEEE Symposium Series on Computational Intelligence, SSCI 2023

作者： Gharavian, Vida Khosrowshahli, Rasa Mahmoud, Qusay H. Makrehchi, Masoud Rahnamayan, Shahryar Lab Ontario Tech University Department of Electrical Computer and Software Engineering OshawaON Canada Brock University Engineering Department St. CatharinesON Canada

ISBN: (纸本)9781665430654

Wireless Sensor Networks (WSNs) are rapidly employed in many applications due to highly demanded autonomous systems. These networks are of immense importance due to their ability to collect data from remote and challenging environments, their impact on various sectors like healthcare, agriculture, industry, environment, and their role in enabling smart technologies for a sustainable, secure, and connected future. Nevertheless, these systems can be attacked by adversaries. Usually, the WSNs are designed with lightweight sensor nodes with limited computation and memory resources. Therefore, employing a firewall system on every sensor node is unacceptable. This paper tackled this problem with a very lightweight Graph Neural Network-based model. The conducted experiment performed in this work demonstrates promising attack-type detection by our proposed approach to the WSN-DS dataset. In this article, our proposed method is compared with other the-state-of-the-art works, and we could discover all Blackhole attacks, one of the most common Denial-of-Service attacks. © 2023 IEEE.

关键词： Denial of Service DoS Graph Neural Network GraphSAGE Intrusion Detection LEACH Wireless Sensor Network WSN

来源：评论

学校读者我要写书评

暂无评论

Predictive Modeling of Students' Stress Levels Using Machine Learning Algorithm 4

Predictive Modeling of Students' Stress Levels Using Machine...

引用

4th International Conference on Emerging Trends in Networks and computer Communications, ETNCC 2024

作者： Ali, Hassan Amin, M. Hamza Rehman, Attique Ur Javaid, Sabeen Ali, Tahir Muhammad Mir, Azka University of Sialkot Department of Software Engineering Sialkot Pakistan GULF University for Sciences and Technology Department of Computer Science Mubarak Al-Abdullah Kuwait

ISBN: (纸本)9798350353266

The increasing prevalence of stress among university students has raised concern about its impact on academic performance and overall well-being. This conference paper explores the applications of machine learning algorithms to predict and analyze the student's stress level. A comprehensive dataset encompassing various stress-related factors, including academic workload, social interactions, and lifestyle, was collected from Kaggle which consists of 1100 instances and 19 attributes. Our methodology places a premium on human connection, integration of qualitative data to refine the accuracy of stress predictions. By seamlessly merging the capability of machine learning with an empathetic understanding of the human experience, our approach strives to pave the way for a more holistic and personalized educational ecosystem. The insights derived from this study aspire to guide educators, administrators, and policy makers in crafting nurturing environments that empower students to excel both academically and emotionally. Embarking on a journey to unravel the myriad factors influencing students' stress, we consider academic pressures, social dynamics, and personal experiences. Harnessing machine learning algorithms such as decision trees, neural networks, we navigate expansive student populations. Acknowledging the nuanced nature of stress, our study integrates a compassionate approach with cutting-edge technology. The study employs state-of-the-art machine learning techniques using various classification algorithms, to build predictive models for assessing stress levels out of which random forest gives 97.45% accuracy. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Variational Auto-Encoder Model and Federated Approach for Non-Intrusive Load Monitoring in Smart Homes 28

Variational Auto-Encoder Model and Federated Approach for No...

引用

28th IEEE Symposium on computers and Communications, ISCC 2023

作者： Kaspour, Shamisa Yassine, Abdulsalam Department of Computer Science Thunder Bay Canada Lakehead University Department of Software Engineering Thunder Bay Canada

ISBN: (纸本)9798350300482

Non-Intrusive Load Monitoring (NILM) is a technique used for identifying individual appliances' energy consumption from a household's total power usage. This study examines a novel energy disaggregation model called Variational Auto-Encoder (VAE) with Federated Learning (FL). Specifically, VAE has a complex structure that resolves the issues in Short Sequence-to-Point (Short S2P) with fewer samples as input windows for each appliance. Short S2P cannot be generalized and might confront some challenges while disaggregating multi-state appliances. To this end, we examine a series of experiments using a real-life dataset of appliance-level power from the UK: UK-DALE. We also investigate additional protection of model parameters using Differential Privacy (DP). The findings show that FL with the VAE model achieves comparable performance to its centralized counterpart and improves all the metrics significantly compared to the Short S2P model. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Multi-Objective Decision Making System Based on Spatial -Temporal Logics 24

Multi-Objective Decision Making System Based on Spatial -Tem...

引用

24th International Conference on Control Systems and computer Science, CSCS 2023

作者： Ababii, Victor Sudacevschi, Viorica Turcan, Ana Melnic, Radu Carbune, Viorel Cojuhari, Irina Technical University of Moldova Department of Computer Sciences and Systems Engineering Chisinau Moldova Technical University of Moldova Department of Software Engineering and Automatics Chisinau Moldova

ISBN: (纸本)9798350313390

Multi-objective decision making systems based on Spatial-Temporal logics presents a class of computational systems based on Artificial Intelligence in which spatial location and time evolution of processes (events) are taken into account. Multi-objective decision making systems showcase a distributed computing architecture based on Agents (Computational Nodes) that form a network for data processing. The functionality of the decision making system is based on events that are spatially and temporally localized. The set of Agents, based on the events, calculates its influence coefficient on the decision taken by the Agent. An Agent can generate both decisions, which lead to personal qualitative evolution, and events that will be processed by other Agents. This work elaborates the following: the multi-objective decision system diagram, the communication model between Agents, the set of operators for spatial-temporal logic, and the model for optimal solution search in event space. © 2023 IEEE.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Enhancing Real-World Facial Expression Recognition: A Deep Learning Approach based on Attention Mechanisms 3

Enhancing Real-World Facial Expression Recognition: A Deep L...

引用

3rd International Conference on computer Science, Electronic Information engineering and Intelligent Control Technology, CEI 2023

作者： Wu, Kai Chen, Zhuang Chongqing University of Posts and Telecommunications Department of Software Engineering Chongqing400065 China Chongqing University of Technology Department of Computer Science and Engineering Chongqing China

ISBN: (纸本)9798350372052

Occlusion and pose variation are the biggest interference factors in real-world face expression recognition (FER). Faces are likely to be occluded by items such as sunglasses, ducktails, and scarves, which can significantly change the visual appearance of the face, while varied poses can lead to partial information loss and inaccurate localization. In this paper, we propose a new deep learning network model for facial expression recognition, firstly, an attention mechanism module is added on top of the residual neural network, which includes spatial and channel attention modules for suppressing unimportant feature information and highlight the key information. Second, a regional attention mechanism is added for adaptively extracting feature vectors from localized unobstructed facial regions, which are combined with global feature information to learn more comprehensive and detailed potential key information. The proposed method achieved good results on two facial expression datasets, FER2013 and RAF-DB, obtaining an accuracy of 74.37 and 89.12, respectively. © 2023 IEEE.

关键词： Face recognition

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：