检索结果-内蒙古大学图书馆

An intelligent hybrid model approach for predictive maintenance of tool wear using machine learning techniques

International Journal of Vehicle Information and Communication Systems 2023年第4期8卷 398-412页

作者： Mohan, Soorya Prabha Nirmala, S. Jaya Department of Computer Science and Engineering National Institute of Technology Tamil Nadu Trichy India

Machine uptime is highly important as the repairing time takes longer which affects the production and the manufacturing industry focus on new ways of being competitive. Manufacturing and assembly parts of the machine are the key component for ensuring machine uptime. Maintenance of these components plays a major role in ensuring the key component health and is an ongoing process. For this, predictive maintenance is the commonly used approach, based on machine running conditions and tool information in production environment. For the proposed methodology, data has been collected from a well-reputed machinery manufacturer. This paper presents the hybrid model, which predicts the machine failure based on multiple ensembling techniques followed by the stacking approach, which performs the k-fold cross-validation and leads to results that provide good accuracy and less false alarms. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Maintenance

来源：评论

学校读者我要写书评

暂无评论

A New Structure of Symmetric/Asymmetric Single Phase Multilevel Inverter to Reduce the Number of Switches and Total Blocking Voltage 9

A New Structure of Symmetric/Asymmetric Single Phase Multile...

引用

9th International Conference on Technology and Energy Management, ICTEM 2024

作者： Seifi, Ali Nadermohammadi, Ali Hosseini, Seyed-Hossein Hagh, Mehrdad Tarafdar Hosseinpour, Majid University of Tabriz Department of Electrical Engineering Ardabil Iran University of Mohaghegh Ardabili Department of Electrical and Computer Engineering Ardabil Iran

ISBN: (纸本)9798350329797

In this research, a novel multilevel inverter is introduced that aims to enhance voltage and power quality by minimizing the number of switches needed at certain voltage levels. The recommended inverter incorporates power-related electrical elements like switches and diodes, along with the inclusion of DC inputs. The design focuses on a fundamental cell consisting of a direct current source, power switch, and diode. The suggested construction introduces an additional switch to the base configuration of the multilevel inverter, resulting in the addition of two levels to the output levels. The addition of two levels to the output voltage is accomplished by including just one additional switch, leading to a decreased number of IGBTs in comparison to conventional structures and comparable investigations. An extensive analysis is performed to compare the suggested structure with the traditional topology and comparable arrangements, taking into account aspects such as the quantity of switches, the quantity of drivers, and overall blocking voltage. The performance characteristics of the suggested configuration are shown via the use of a single-phase, 9-level inverter. The structure is comprehensively analyzed using simulations in the Matlab/Simulink, and practical tests are conducted using a laboratory prototype to verify its real-world viability. The suggested structure is notable for its benefits, including its streamlined design, improved effectiveness, and greater power and voltage characteristics. © 2024 IEEE.

关键词： PIN diodes

来源：评论

学校读者我要写书评

暂无评论

Quantized Spiking Neural Networks on FPGA: An Application to Retinal Prosthetics

Quantized Spiking Neural Networks on FPGA: An Application to...

引用

2023 IEEE Biomedical Circuits and Systems Conference, BioCAS 2023

作者： Besrour, Marwan Bouchard, Samuel Mohanty, Romit Lavoie, Jacob Omrani, Takwa Gouin-Ferland, Berhtie Koua, Konin Therrien, Audrey C. Boukadoum, Mounir Fontaine, Rejean Université de Sherbrooke Department of Electrical Engineering and Computer Engineering Canada Université du Québec À Montréal Department of Computer Science Canada

ISBN: (纸本)9798350300260

We present an embedded Spiking Neural Network (SNN) in a Field-Programmable Gate Array (FPGA) for retinal prosthetic applications. Our primary goal is to minimize resource utilization, making the solution suitable for edge AI systems with neural-machine interfaces. The proposed quantized SNN minimizes computational resources for a power-efficient system. The PRANAS open-source software, which emulates retinal ganglion cells (RGCs), is used to generate a spiking dataset from popular MNIST database and the developed SNN is trained on it by the error backpropagation and surrogate gradient technique. We subsequently applied weight quantization on the trained SNN's weights, using both 4-bit and 8-bit precision, to create a compact and power-efficient version of the SNN. The resulting accuracies after the 4-bit and 8-bit post-training quantizations are 83.2% and 87.2%, respectively. They show the potential of FPGA-based implementations of quantized SNNs to offer high-performance and energy-efficient solutions for retinal implants, thus answering their inherent hard power and area constraints. © 2023 IEEE.

关键词： FPGA leaky-integrate-and-fire LIF machine learning Neuromorphic SNN spiking neural networks

来源：评论

学校读者我要写书评

暂无评论

PDC-ViT: source camera identification using pixel difference convolution and vision transformer

引用

Neural Computing and Applications 2025年第9期37卷 6933-6949页

作者： Elharrouss, Omar Akbari, Younes Almadeed, Noor Al-Maadeed, Somaya Khelifi, Fouad Bouridane, Ahmed Department of Computer Science and Software Engineering United Arab Emirates University Al Ain United Arab Emirates Department of Computer Science and Engineering Qatar University Doha Qatar Northumbria University Newcastle Newcastle upon Tyne United Kingdom University of Sharjah Sharjah United Arab Emirates

Source camera identification has emerged as a vital solution to unlock incidents involving critical cases like terrorism, violence, and other criminal activities. The ability to trace the origin of an image/video can aid law enforcement agencies in gathering evidence and constructing the timeline of events. Moreover, identifying the owner of a certain device narrows down the area of search in a criminal investigation where smartphone devices are involved. This paper proposes a new pixel-based method for source camera identification, integrating Pixel Difference Convolution (PDC) with a Vision Transformer network (ViT), and named PDC-ViT. While the PDC acts as the backbone for feature extraction by exploiting Angular PDC (APDC) and Radial PDC (RPDC). These techniques enhance the capability to capture subtle variations in pixel information, which are crucial for distinguishing between different source cameras. The second part of the methodology focuses on classification, which is based on a Vision Transformer network. Unlike traditional methods that utilize image patches directly for training the classification network, the proposed approach uniquely inputs PDC features into the Vision Transformer network. To demonstrate the effectiveness of the PDC-ViT approach, it has been assessed on five different datasets, which include various image contents and video scenes. The method has also been compared with state-of-the-art source camera identification methods. Experimental results demonstrate the effectiveness and superiority of the proposed system in terms of accuracy and robustness when compared to its competitors. For example, our proposed PDC-ViT has achieved an accuracy of 94.30%, 84%, 94.22% and 92.29% using the Vision dataset, Daxing dataset, Socrates dataset and QUFVD dataset, respectively. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

An Empirical Assessment of Botnets and Detection Methods 15

An Empirical Assessment of Botnets and Detection Methods

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Shivanna, D. Pooja, M.R. Department of Computer Science and Engineering Vidyavardhaka College of Engineering Mysuru India

ISBN: (纸本)9798331300579

a botnet is a group of devices which are controlled by the Attacker and used for fraud, scam and *** attacker is also known as *** Bots are used to conduct many activities varies from stealing user credentials, sending spam, stealing data, ransom ware, fraudulently clicking on ads or distributed denial-of-service (DDoS) *** traffic made up by bots is increasing day by day, now nearly a quarter of the whole *** them becomes an uphill task as the time *** developers are adopting new technologies at a rapid rate in an effort to avoid bot *** some bot problems are industry-specific, researchers found out that bot problems run across all *** top five industries which are targeted with bad bot traffic are the financial, education, IT & services industries, marketplaces, and *** assessment paper briefs the different botnet detection methods that exist at *** main aim of this paper is to give the insight knowledge of various bots and their implications on the present cyber world and also shedding some light on the current methods using to detect them... © Grenze Scientific Society, 2024.

关键词： Bot (Internet)

来源：评论

学校读者我要写书评

暂无评论

SUDM-SP:A method for discovering trajectory similar users based on semantic privacy

引用

High-Confidence Computing 2023年第3期3卷 68-80页

作者： Weiqi Zhang Guisheng Yin Bingyi Xie College of Computer Science and Technology Harbin Engineering UniversityHarbin 150009China Department of Computer Science Georgia State UniversityGA 30303USA

With intelligent terminal devices’widespread adoption and global positioning systems’advancement,Location-based Social Networking Services(LbSNs)have gained considerable *** recommendation mechanism,which revolves around identifying similar users,holds significant importance in *** order to enhance user experience,LbSNs heavily rely on accurate *** mining and analyzing users who exhibit similar behavioral patterns to the target user,LbSNs can offer personalized services that cater to individual ***,trajectory data,a form encompassing various sensitive attributes,pose privacy *** disclosure of users’precise trajectory information can have severe consequences,potentially impacting their daily ***,this paper proposes the Similar User Discovery Method based on Semantic Privacy(SUDM-SP)for trajectory *** approach involves employing a model that generates noise trajectories,maximizing expected noise to preserve the privacy of the original *** users are then identified based on the published noise trajectory ***-SP consists of two key ***,a puppet noise location,exhibiting the highest semantic expectation with the original location,is generated to derive noisesuppressed trajectory ***,a mechanism based on semantic and geographical distance is employed to cluster highly similar users into communities,facilitating the discovery of noise trajectory similarity among *** trials conducted using real datasets,the effectiveness of SUDM-SP,as a recommendation service ensuring user privacy protection is substantiated.

关键词： Differential privacy LbSNs Semantic trajectory Recommendation

来源：评论

学校读者我要写书评

暂无评论

A Survey on Tea Plant Disease Prediction for Blister Blight 2

A Survey on Tea Plant Disease Prediction for Blister Blight

引用

2nd International Conference on Emerging Trends in Information Technology and engineering, ic-ETITE 2024

作者： Gowthami, M. Nila Baaradhy, I. Nithin, V. Subhash, J. Easwari Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350328202

This survey seeks to give a general overview of the current state of Tea Plant disease prediction for the Blister Blight disease. To construct this system, IoT and machine learning approaches are applied. The goal of the research is to improve sustainable farming methods through early detection and prevention. The climatic conditions of the farmlands are analysed with a set of sensors. Real time input from the IoT setup is processed by the Machine Learning algorithms for detection of the Blister Blight disease. © 2024 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Unveiling Marketing Potential: Harnessing Advanced Analytics and Machine Learning for Gold Membership Strategy Optimization in a Superstore

引用

SN computer science 2024年第4期5卷 374页

作者： Singh Mahala, Vikas Ranveer Garg, Neeraj Saxena, D. Kumar, Rajesh Department of Electrical Engineering Swami Keshvanand Institute of Technology Management & amp Department of Computer Science and Engineering Swami Keshvanand Institute of Technology Management & amp Department of Electrical Engineering Malaviya National Institute of Technology Rajasthan Jaipur 302017 India

This research paper presents a comprehensive case study conducted in a superstore, introducing a novel gold membership offer and employing sophisticated analytics and machine learning methodologies to identify potential customers. The primary objective of this study is to explore available data to discern the factors influencing customers’ responses to a new supermarket offering. Subsequently, a predictive model is developed to accurately gauge the likelihood of a favorable customer response. In pursuit of enhancing marketing strategies and bolstering sales, this study employs a suite of machine learning techniques, including decision trees, support vector machines, random forests, and XGBoost. Furthermore, the study incorporates metaheuristic optimization algorithms such as grey wolf optimization, slime mold algorithm, multi-verse optimizer, and particle swarm optimization to fine-tune hyperparameters of the machine learning models. These optimization algorithms serve as effective search mechanisms, facilitating the identification of optimal solutions and significantly improving classification performance in the context of the complex superstore problem. The research findings highlight the substantial impact of the metaheuristic strategy, specifically grey wolf optimization, on the performance of all machine learning models. Notably, the random forest model achieved the highest accuracy of 95% with the application of grey wolf optimization. Moreover, the decision tree model demonstrated remarkable improvement in accuracy following hyperparameter tuning with grey wolf optimization. Collectively, these results underscore the critical role of metaheuristic optimization in enhancing the performance of machine learning models for marketing strategies in the superstore industry. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Decision tree GWO Hyperparameter MVO Optimization PSO Random Forest SMA Supermarket SVM XGBoost

来源：评论

学校读者我要写书评

暂无评论

Multi-Modal Visual Features Perception Technology for Internet of Vehicles (IoV) 6

Multi-Modal Visual Features Perception Technology for Intern...

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Mahendran, Rakesh Kumar Thiyagarajan, Abirami Angelin Gracia, A. Kumar, P. Rajalakshmi Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350306613

In the realm of the Internet of Vehicles (IoV), this study addresses a fundamental question that causes the emerging trends within the IoV landscape, Previous research in IoV has laid the groundwork but is limited by conventional single-modal methods. This study advances our understanding by incorporating innovative Multi-Modal Visual Features Perception Technology by inputting various images from LiDAR cameras, Radar and processing using ConvolutionalNeural Networks (CNN), alongside single-modal approaches. Through this research, we uncover hidden facets of IoV, promising transformative insights. The implementation of Multi-Modal Visual Features Perception Technology using deep learning, along with single-modal methods, holds the potential to revolutionize vehicular connectivity and shape a more connected and secure future. © 2024 IEEE.

关键词： Advanced neural network techniques Convolutional Neural Networks LiDAR Multimodal Scene perception Self-driving technology sensor fusion

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Solid Waste Classification and Monitoring Alert System using Deep Learning

An Intelligent Solid Waste Classification and Monitoring Ale...

引用

2024 International Conference on Integration of Emerging Technologies for the Digital World, ICIETDW 2024

作者： Selvi, S. Elamathy, G.O. Kirana, B. Swetha, S. Government College of Engineering Department of Computer Science and Engineering Bargur India

ISBN: (纸本)9798331541170

Efficient waste control is critical for India's smart cities. This study uses Convolutional Neural network (CNN) technology to classify and reveal waste, integrating sensors, Internet of Things (IoT) gadgets, and system studying to promote sustainability. Also targets to deal with inefficient waste management by using CNN technology for effective garbage classification and tracking. The device integrates sensors, IoT devices, and machine learning algorithms to sort and track waste, promoting sustainable practices and environmental conservation. The proposed system uses an ArduinoMKR1000 microcontroller, ultrasonic sensors, and jumper wires to monitor garbage bin levels and classify waste into biodegradable and non-biodegradable classes the usage of CNN algorithms. The system employs the DenseNet121 model for image type and makes use of ThingSpeak for real-time tracking and notifications. The experimental consequences display an accuracy of 95% for the DenseNet121 version, outperforming traditional CNN and ResNet-50 architectures. The proposed device's overall performance surpasses that of different models because of its dense connectivity and efficient feature map creation. Real-time monitoring and data analytics provide precious insights for waste reduction and sustainability tasks, contributing to cleaner, greener, and more sustainable environments for future generations. © 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

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：