检索结果-内蒙古大学图书馆

36th International Florida Artificial Intelligence Research Society Conference, FLAIRS-36 2023

作者： Borowski, Richard Choi, Arthur Computer Science Department College of Computing and Software Engineering Kennesaw State University United States

A neuron with binary inputs and a binary output represents a Boolean function. Our goal is to extract this Boolean function into a tractable representation that will facilitate the explanation and formal verification of a neuron's behavior. Unfortunately, extracting a neuron's Boolean function is in general an NP-hard problem. However, it was recently shown that prime im-plicants of this Boolean function can be enumerated efficiently, with only polynomial time delay. Building on this result, we propose a best-first search algorithm that is able to incrementally tighten inner and outer bounds of a neuron's Boolean function. These bounds correspond to truncated prime-implicant covers of the Boolean function. We provide two case studies that highlight our ability to bound the behavior of a neuron. © 2023 by the authors. All rights reserved.

关键词： Boolean functions

来源：评论

学校读者我要写书评

暂无评论

A Cloud-Based 3D Digital Twin for Arabic Sign Language Alphabet Using Machine Learning Object Detection Model 15

A Cloud-Based 3D Digital Twin for Arabic Sign Language Alpha...

引用

15th International Conference on Innovations in Information Technology, IIT 2023

作者： Abduljabbar, Mohammed Gochoo, Munkhjargal Sultan, Mohamad T. Batnasan, Ganzorig Otgonbold, Munkh-Erdene Berengueres, Jose Alnajjar, Fady Rasheed, Ashika Abdul Alshamsi, Ahmed Alsaedi, Naser United Arab Emirates University Department of Computer Science and Software Engineering United Arab Emirates

ISBN: (纸本)9798350382396

People with hearing loss or hard hearing struggle with daily life activities as sign language is not widely known by the public. There are many attempts to use technology to help assist hearing loss individuals. However, most proposed solutions are standalone applications or require special hardware like a wearable glove. Our goal is to leverage cloud computing and artificial intelligence (AI) to provide a solution that is portable and does not require any special hardware. We created a lightweight 3D model and rendered it on the browser along with another lightweight object detection model for Arabic Sign Language (ArSL) for real-Time detection. Our contribution is primarily based on integrating our novel functional lightweight 3D avatar model and a lightweight ArSL alphabet detection model, which is trained on public ArSL21L dataset, that are suitable to be given as a cloud service. Prototypes of the 3D digital twin avatar model and AI model are publicly offered for the research community on GitHub. We will be working on a full-scale real-Time cloud-based communication system in ArSL. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Sri Potha: A Voice-Based Sinhala Document Maker

Sri Potha: A Voice-Based Sinhala Document Maker

引用

2023 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems, ICSES 2023

作者： De Silva, D.I. Piyadali, D.S.P. Dilrukshi, M.G.S. Sri Lanka Institute of Information Technology Department of Computer Science and Software Engineering Sri Lanka

ISBN: (纸本)9798350319200

This study discusses the development of a web application aimed at facilitating Sinhala document creation, with a specific emphasis on Sinhala voice-to-text conversion and the handling of Sinhala commands through vocal input. Leveraging machine learning techniques, including convolutional neural networks and Natural Language Processing, the application's core features were established. Extensive research was conducted to tailor the application's content to the needs of its primary users, ensuring maximum effectiveness. The user-friendly interfaces of the web application are designed for clarity, simplicity, and consistency. The primary objective of this research is to comprehensively analyze the implementation of Sinhala voice-to-text conversion and Sinhala command handling systems. These systems are primarily designed to benefit diverse users, including journalists, content writers, and differently abled individuals with verbal abilities, by enhancing the efficiency of creating Sinhala documents. Through a detailed exploration of the research methodology, this study offers insight into the development process of the web-based system. The outcomes of the linguistic model training, presented within the study, reveal achievements and advancements that address the limitations inherent in existing solutions. Key findings from this research demonstrate the successful functionality of the Sinhala voice-to-text converter and the efficacy of the Sinhala command handler. The voice-to-text conversion system achieved an impressive accuracy rate of over 80%, while the Sinhala command handler exhibited an accuracy of approximately 80%. Moreover, this research envisions potential applications that extend beyond document creation. The technology showcased in the web application holds promise for broader language-based applications, impacting education, accessibility, and communication across the native Sinhala-speaking community in Sri Lanka. In summary, this research showcases th

关键词： convolutional neural network machine learning natural language processing

来源：评论

学校读者我要写书评

暂无评论

Enhancing Deep Convolutional Neural Networks to Detect Pedestrians Effectively in Various Urban Environments 6

Enhancing Deep Convolutional Neural Networks to Detect Pedes...

引用

6th International Conference on Electrical engineering and Information and Communication Technology, ICEEICT 2024

作者： Hasan, Md. Mehedi Aman, Nakib Jahan, Md. Wadud Software Engineer i Developer EXperience Hub Rajshahi Bangladesh Pabna University of Science and Technology Department of Computer Science and Engineering Pabna Bangladesh

ISBN: (纸本)9798350385779

In computer vision, accurately identifying pedestrians is a crucial difficulty, especially in applications where safety is of the highest priority, such autonomous driving and urban planning. Deep learning has made progress, but challenges such changing conditions, barriers, and processing power requirements still exist. In this paper, we present an improved deep convolutional neural network (DCNN) designed with pedestrian identification in mind. Improving accuracy in various settings while maximizing processing efficiency is our goal. We carefully prepare the data and create a novel DCNN architecture that can handle a variety of pedestrian appearances and urban surroundings by utilizing well-known datasets such as Penn-Fudan and PnPLO. Our model performs exceptionally well in terms of accuracy, speed, and durability, according to a thorough evaluation that compares it to recognized benchmarks. In addition to making important advances, our study has shown novel approaches to enhancing performance in challenging environments, such as the analysis of transfer learning and the deployment of real-time systems. This work paves the way for future developments in intelligent, self-governing systems and is a maj or leap forward in pedestrian detection technology. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Tomato Leaf Diseases Prediction Using Deep Learning Algorithms

Tomato Leaf Diseases Prediction Using Deep Learning Algorith...

引用

2024 International Conference on Advances in Data engineering and Intelligent Computing Systems, ADICS 2024

作者： Ananthi, P. Devi, K. Nirmala Gopinath, D. Shanmugapriya, P. Gowtham, S. Kongu Engineering College Department Of Computer Technology Tamilnadu Perundurai India Bangaluru India Kristu Jayanti College Department Of Computer Science Bangaluru India Software Systems Kongu Engineering College Erode India

ISBN: (纸本)9798350364828

In the realm of agriculture, infections on tomato leaves pose a worldwide danger to established tomato production, impacting a large number of farmers worldwide. To ensure healthy tomato plant growth and food security for the world's expanding population, early disease detection, treatment, and solution identification are essential. The use of computer-assisted technologies for disease detection in plant leaf is prevalent in modern agriculture. The goal was to develop an effective deep learning model for the detection of diseases in tomato plant leaf. To address memory and computational resource constraints, the proposed model incorporates a lightweight 'Region Convolutional Neural Network (R-CNN)' head and conducted experiments with various neural network architectures, including Densenet-169, optimized Mobilenet-V2 and Resnet-50, to enhance detection accuracy and computational efficiency. The research extends the capabilities of these neural network architectures by optimizing anchor proportions in the Conventional Neural Network (CNN) and modifying feature extraction topologies. These optimizations resulted in improved detection accuracy and computational efficiency and compared proposed technique against cutting-edge models to evaluate its viability and robustness. The outcomes of the research were promising. Impressive outcomes were obtained by the proposed approach regarding accuracy, F1-score, recall, and precision. © 2024 IEEE.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Spatiotemporal crowds features extraction of infrared images using neural network

引用

Journal of Ambient Intelligence and Humanized Computing 2024年第4期15卷 2543-2556页

作者： Al-Oraiqat, Anas M. Drieiev, Oleksandr Drieieva, Hanna Meleshko, Yelyzaveta AlRawashdeh, Hazim Al-Oraiqat, Karim A. Hasan, Yassin M. Y. Maricar, Noor Khan, Sheroz Department of Computer Science College of Engineering and Information Technology Onaizah Colleges Qassim Saudi Arabia Department of Cybersecurity and Software Central Ukrainian National Technical University Kropyvnytskyi Ukraine Department of Artificial Intelligence Faculty of Information Technology Zarqa University Zarqa Jordan Department of Electrical Engineering College of Engineering and Information Technology Onaizah Colleges Qassim Saudi Arabia

Crowds can lead up to severe disasterous consequences resulting in fatalities. Videos obtained through public cameras or captured by drones flying overhead can be processed with artificial intelligence-based crowd analysis systems. Being a hot area of research over the past few years, the goal is not only to identify the presence of crowds but also to predict the probability of crowd-formation in order to issue timely warnings and preventive measures. Such systems will significantly reduce the probablity of the potential disasters. Developing effective systems is a challenging task, especially due to factors such as naturally occuring diverse conditions, variations in people or background pixel areas, noise, behaviors of individuals, relative amounts/distributions/directions of crowd movements, and crowd building reasons. This paper proposes an infrared video processing system based on U-Net convolutional neural network for crowd monitoring in infrared video frames to help estimate the people crowd with normal or abnormal trends. The proposed U-Net architecture aims to efficiently extract crowd features, achieve sufficient people marking-up accuracy, competitively with optimal network configurations in terms of the depth and number of filters to consequently minimise the number of coefficients. For further faster processing, hardware resources/implementation area savings, and lower power, the optimized network coefficients measured are represented in Canonic-Signed Digit with minimal number of nonzero (± 1) digits, minimizing the number of underlying shift-add/subtract operations of all multipliers. The achieved significantly reduced computational cost makes the proposed U-Net effectively suitable for resource-constrained and low power applications. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Hypergraph Transformer With Adaptive EdgeConv for Skeleton-based Action Recognition 7

A Hypergraph Transformer With Adaptive EdgeConv for Skeleton...

引用

7th International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2024

作者： Zhou, Fengyu Liu, Jun Zhang, Dazhi Huang, Zengxi School of Computer Science and Software Engineering Xihua University Chengdu China Sichuan Police College Criminal Science and Technolog Department Luzhou China

ISBN: (纸本)9798350350890

Skeleton-based action recognition is a crucial task in computer vision, aiming to perform pattern recognition on a skeleton sequence to identify the actions represented by the skeleton's semantics. However, current mainstream skeleton-based methods primarily rely on global edge information. Unimportant edges may influence this approach, limiting the model's performance in action recognition tasks. Additionally, because these methods treat all information equally, they may lead to resource wastage. This paper proposes a Hypergraph Transformer-based action recognition method using Adaptive EdgeConv Module(AEM) to address these issues effectively. In this method, we introduce Adaptive Edge Graph Convolution Networks(AEGCNs), which utilize the AEM to adaptive obtain edge sets and edge set weights, thereby adaptive identifying key edge sets. With this information, the model can dynamically adjust the importance of different edge sets based on various samples. This approach improves the accuracy of action recognition and reduces the loss of computational resources. Our experimental results on two challenging benchmark datasets (i.e., NTU RGB+D and NTU RGB+D 120) show that the proposed method is comparable to state-of-the-art methods. © 2024 IEEE.

关键词： Musculoskeletal system

来源：评论

学校读者我要写书评

暂无评论

Cryptcoder: An Automatic Code Generator for Cryptographic Tasks in Ethereum Smart Contracts 31

Cryptcoder: An Automatic Code Generator for Cryptographic Ta...

引用

31st IEEE International Conference on software Analysis, Evolution and Reengineering, SANER 2024

作者： Xia, Libin Zhang, Jiashuo Wang, Che Tan, Zezhong Gao, Jianbo Guan, Zhi Chen, Zhong Peking University MoE Key Laboratory of High Confidence Software Technologies Beijing China School of Computer Science Peking University Beijing China Peking University National Engineering Research Center for Software Engineering Beijing China

ISBN: (纸本)9798350330663

Cryptographic APIs provided by Ethereum are widely adopted in decentralized applications (DApps) for cryptographic operations. However, developers who lack expertise in cryptography frequently encounter difficulties when working with low-level cryptographic APIs, thereby producing insecure code. To address this issue, we introduce Cryptcoder, an automatic code generator designed to bridge the gap between low-level cryptographic APIs and high-level cryptographic tasks in Ethereum. The fundamental component of Cryptcoder is Cryptlang, a Solidity-compatible domain-specific language (DSL) designed for cryptographic tasks. Developers can utilize Cryptlang for the straightforward and secure implementation of cryptographic tasks, such as signatures and commitments, and employ Cryptcoder for the automatic conversion into Solidity code. The evaluation of Cryptcoder demonstrates both its functionality in generating Solidity code and an acceptable overhead, evidenced by a mere 4% average increase in gas costs compared to the reference code. A demonstration video of Cryptcoder is available at https://***/AxhCdGiu7dw. © 2024 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Gaussian Differential Privacy Integrated Machine Learning Model for Industrial Internet of Things

引用

SN computer science 2023年第5期4卷 454页

作者： Lazar, Arokia Jesu Prabhu Soundararaj, Sivaprakash Sonthi, Vijaya Krishna Palanisamy, Vishnu Raja Subramaniyan, Vanithamani Sengan, Sudhakar Department of Computer Science and Engineering AI & ML CMR Institute of Technology Telangana Hyderabad 501401 India Department of Software Systems School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore 632014 India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Vaddeswaram 522502 India Department of Computer Science and Engineering Kongu Engineering College Tamil Nadu Perundurai 638060 India Master of Computer Applications M. Kumarasamy College of Engineering Tamil Nadu Karur 638060 India Department of Computer Science and Engineering PSN College of Engineering and Technology Tamil Nadu Tirunelveli 627152 India

Agriculture, energy, mining, healthcare, and transportation are a few of the top industries transformed by the industrial internet of things (IIoT). Industry 4.0 mainly relies on machine learning (ML) to use the vast interconnectedness and large amounts of IIoT data that IIoT primarily drives. ML approaches are trained on confidential data generated by the IIoT environment often exposes privacy to adversarial assaults. Blockchain-based ML is established in the proposed secured IIoT research work to safeguard and enhance privacy. This research proposes a Gaussian differential privacy-integrated machine learning model (GDPIMLM), created for a scalable and controlled IIoT system, fully connected (FC) layer, implemented blockchain benefits to create a privacy-preserving mechanism while considering other limitations and reasonable time. A probability test using two altered Gaussian distributions (GD) forms the basis for defining Gaussian differential privacy (GDP). Due to a central restriction theorem for differential privacy, GDP is the primary privacy term within the family of Federal differential privacy (f-DP) policies. Ethereum is used to implement experimental evaluations. Data collections show that the recommended method enhances digital data privacy with industry-leading security without reducing performance. The performance analysis of the proposed model of secured industrial internet of things (IIoT) is studied using the accuracy of comparison with existing methods and for diverse distributed entities (DE). The proposed secured IIoT attains a minimal time consumption (TC) of 790 Sec—the highest accuracy of 92%, outperforming the existing method. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Blockchain Industry Industry 4.0 Learning Privacy Security Smart objects

来源：评论

学校读者我要写书评

暂无评论

A Review of Colonial Pipeline Ransomware Attack 23

A Review of Colonial Pipeline Ransomware Attack

引用

23rd IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing Workshops, CCGridW 2023

作者： Beerman, Jack Berent, David Falter, Zach Bhunia, Suman Miami University Department of Computer Science and Software Engineering OxfordOH45056 United States

ISBN: (纸本)9798350302080

In April, 2021 a ransomware attack occurred on Colonial Pipeline. The details of this attack point to the hacking group DarkSide taking advantage of the design flaws in the Colonial Pipeline network. After extensive research, the specificity of this attack was related to VPN access through an unused account. In order to regain control of their systems, Colonial Pipeline paid the attackers. This act has only created an incentive for similar attacks across the nation. The events of this attack have impacted both the United States, where the company is located and the world in a very negative way. This paper analyzes the attack with published data and provides a detailed attack methodology. From the attack methodology the focus then shifts into the impacts that an attack of this caliber had, on both the company, the United States, and the world. We then outline possible defense strategies against this type of ransomware attack, analyzing what could have been done to prevent this attack from happening. In addition, we also detail how companies can prevent future attacks of this caliber. Finally we wrap up or findings and detail the key takeaways of the entire attack. © 2023 IEEE.

关键词： Virtual private networks

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：