检索结果-内蒙古大学图书馆

2023 IEEE International Conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering, RMKMATE 2023

作者： Paing, Soe Htun, Yadana Faculty of Computer Systems and Technologies Myanmar Institute of Information Technology Mandalay Myanmar Pyin Oo Lwin Myanmar

ISBN: (纸本)9798350305708

People utilize contemporary tools to edit captured video clips effortlessly. These video clips hold significance as evidence within a legal setting. Consequently, the identification of video tampering becomes crucial during legal proceedings. In order to improve the legal process, this paper presents an inter-frame forgery detection system especially designed for frame deletions. The system initially receives video clips and employs a statistical approach to determine whether they have been tampered with or are authentic. If the video system is determined to be forged, it will proceed with detecting frame deletions or other alterations. This detection process utilizes an unsupervised learning algorithm, specifically the K-mean learning algorithm. By accepting the cross-correlation coefficient value, the detection process categorizes the different types of forgery. The conducted tests on video clips sourced from two data sets: the University of Central Florida (UCF) 101 data sets and Surrey University Library for Forensic Analysis (SULFA) data sets to evaluate the system. The outcomes of the conducted investigation clearly showcase the effectiveness of the suggested system in precisely identifying instances of frame deletion forgery, achieving an impressive success rate of 87%. © 2023 IEEE.

关键词： Forgery types detection frame deletion Inter-frame forgery K mean clustering unsupervised learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Pressure pattern recognition in buildings using an unsupervised machine-learning algorithm

引用

JOURNAL OF WIND ENGINEERING AND INDUSTRIAL AERODYNAMICS 2021年 214卷 104629-104629页

作者： Kim, Bubryur Yuvaraj, N. Tse, K. T. Lee, Dong-Eun Hu, Gang Dong A Univ Dept Architectural Engn Busan 49315 South Korea Dong A Univ Dept ICT Integrated Ocean Smart Cities Engn Busan 49315 South Korea Hong Kong Univ Sci & Technol Dept Civil & Environm Engn Kowloon Clear Water Bay Hong Kong Peoples R China Kyungpook Natl Univ Sch Architecture Civil Environm & Energy Engn 80 Daehak Ro Daegu 41566 South Korea Harbin Inst Technol Sch Civil & Environm Engn Shenzhen 518055 Peoples R China

Owing to its significance in ensuring structural safety and occupant comfort, wind pressure on buildings has attracted the attention of numerous scholars. However, the characteristics of wind pressures are usually complex. This study employs an unsupervised machine-learning algorithm, clustering algorithms, to study wind pressures on buildings. Wind pressures on a single building and two adjacent buildings with different gaps are measured in a wind tunnel, with clustering algorithms applied to cluster different wind pressure patterns. The results show that for the single-building model, the pressure patterns are symmetrical on the side surfaces of the building;for the two-building model with a small gap, a channeling effect can be identified;for the two-building model with a large gap, the pressure patterns shared symmetry with that of the single-building model. Clustering algorithms can recognize unidentified patterns of wind pressures on buildings. This study demonstrates that clustering algorithms are a powerful tool for recognizing patterns hidden in complex pressure fields and flow fields. Therefore, this study proposes a promising machine-learning technique that can perfectly complement traditional building methods using wind engineering.

关键词： unsupervised learning algorithm Wind tunnel testing Pressure pattern Clustering algorithms Pattern recognition Wind pressure

来源：评论

学校读者我要写书评

暂无评论

Channel Parameter and Read Reference Voltages Estimation in 3-D NAND Flash Memory Using unsupervised learning algorithms

引用

IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 2024年第1期43卷 305-318页

作者： Hu, Haihua Han, Guojun Wu, Wenhua Liu, Chang Guangdong Univ Technol Sch Informat Engn Guangzhou 510006 Peoples R China

In 3-D NAND flash memory, the channel is always offset due to the complicated interference from the program/erase (PE), including the data retention and layer interference, so that the channel estimation is desired. However, due to the physical structure of 3-D flash memory, there exists significant variation in channel parameters and inconsistent channel offsets among different wordlines. As a result, the process of estimating channel parameters for each individual wordline typically requires a considerable amount of computational resources, resulting in the high latency of system, which becomes a new challenge. To tackle this problem, two unsupervised learning algorithms are proposed to estimate the channel parameters, based on analyzing the error distribution in 3-D flash memory. To address the longer read latency introduced by the unsupervised learning algorithm for the channel estimation, we further propose a low-latency detection algorithm, which first detects whether the current channel needs to be updated. In the event that an update is required, the algorithm only periodically estimates the channel parameters during system idle times, resulting in a more efficient and streamlined process. Compared to the existing methods, the proposed algorithms can efficiently estimate channel parameters with lower-computational complexity. Moreover, combining with the search algorithm, a correction scheme is proposed to minimize the error between the estimated read reference voltage (RRV) and the optimal RRV in the actual device. Theoretical analysis and simulation results demonstrate that the proposed method can improve the lifetime of flash memory and reduces the number of read retries.

关键词： 3-D NAND flash memory low complexity param-eter estimation read reference voltage (RRV) unsupervised learning algorithm

来源：评论

学校读者我要写书评

暂无评论

unsupervised learning algorithm for Signal Separation

Unsupervised Learning Algorithm for Signal Separation

引用

IEEE Symposium on Computational Intelligence for Multimedia, Signal and Vision Processing (CIMSIVP)

作者： Jacob, Theju Snyder, Wesley North Carolina State Univ Dept Elect & Comp Engn Raleigh NC 27695 USA

ISBN: (纸本)9781479945030

We present a neural network capable of separating inputs in an unsupervised manner. Oja's rule and Self-Organizing map principles are used to construct the network. The network is tested using 1) straight lines 2) MNIST database. The results demonstrate that the network can operate as a general clustering algorithm, with neighboring neurons responding to geometrically similar inputs.

关键词： pattern clustering self-organising feature maps source separation unsupervised learning Oja rule general clustering algorithm neural network neurons self-organizing map principles signal separation unsupervised learning algorithm Clustering algorithms Hebbian theory Lattices Neurons Training

来源：评论

学校读者我要写书评

暂无评论

An unsupervised learning algorithm for the Classification of the Protection Device in the Fault Diagnosis System 6

An Unsupervised Learning Algorithm for the Classification of...

引用

China International Conference on Electricity Distribution (CICED)

作者： Li, Bin Guo, Yajuan Wu, Yi Chen, Jinming Yuan, Yubo Zhang, Xiaoyi State Grid Jiangsu Elect Power Res Inst Nanjing Jiangsu Peoples R China State Grid Jiangsu Elect Power Co Nanjing Jiangsu Peoples R China

ISBN: (纸本)9781479941254

Power protection devices achieve a rapid removal of the grid accident, but the numerous applications of the devices had brought data disaster for the fault diagnosis information system. It costs a great deal of efforts to ensure that the information uploaded by the protection devices correspond to the function of the devices. This paper presents an unsupervised learning algorithm for the classification of the protection device to facilitate the fault diagnosis information system to locate accurately the event reports of every protection device. The algorithm classifies the protection devices without samples or with small number of samples according to the automation relaying settings. The classification of the protection devices is not just separating the devices according to the types of different companies, but also differentiates these devices with the same type from the same company but different functions. There are two innovations in the proposed unsupervised learning algorithm for the classification of the protection device. Firstly, the automation relaying settings can solve the classification of the protection device in essence and eliminate mistakes from the source. The mistakes are usually caused by the classification of the device's name in manual mode. Secondly, addressing to the characteristics that the relaying settings of the protection devices have uncertain entries under different functions, the algorithm realizes the classification from massive devices.

关键词： Abstracts Automation Monitoring Niobium Visualization classification malfunction protection device unsupervised learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Application of unsupervised learning algorithms in Environmental Factor Correlation Analysis and Model Construction

引用

Procedia Computer Science 2025年 261卷 148-157页

作者： Yu Zhao Ting Zhang Xinjiang Shengyuan Xianghe Environmental Protection Engineering Co. Ltd. Urumqi Xinjiang China

As environmental protection issues become more serious, scientists and policy makers are in urgent need of efficient tools to analyze and manage environmental data. This study explores the application of unsupervised learning algorithms in correlation analysis and model building, with a special focus on the effectiveness of clustering, dimensionality reduction and density estimation algorithms in processing and interpreting complex environmental data. By applying K-mean clustering, Principal Component Analysis (PCA), self-organizing mapping, and Density-Based Spatial Clustering of Applications with Noise (DBSCAN), we identified air and water pollution sources, assessed the environmental quality of multidimensional environmental data, monitored the health of ecosystems, and evaluated environmental risk areas. Higher concentrations of PM2.5 (2.5-micrometer Particulate Matter), NO2, and SO2 in the industrial areas (monitoring points 1, 3, 5, 7, and 10) indicate that there are serious air pollution problems in these areas. The results show that the unsupervised learning algorithm can effectively extract key information from environmental monitoring data to support the formulation and implementation of environmental policies. This study not only extends the application of unsupervised learning in the field of environmental sciences, but also provides a reliable data-driven method for environmental management, which helps to improve the science and effectiveness of environmental protection and sustainable development.

关键词： unsupervised learning algorithm Environmental Factor Correlation Analysis Model Construction Density-Based Clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

unsupervised learning for lithium-ion batteries fault diagnosis and thermal runaway early warning in real-world electric vehicles

引用

JOURNAL OF ENERGY STORAGE 2025年 109卷

作者： Yu, Quanqing Yang, Yu Tang, Aihua Wu, Zikang Xu, Yuchen Shen, Weixiang Zhou, Feng Harbin Inst Technol Sch Automot Engn Weihai 264209 Shandong Peoples R China Chongqing Univ Technol Key Lab Adv Mfg Technol Automobile Parts Minist Educ Chongqing 400054 Peoples R China Swinburne Univ Technol Sch Sci Comp & Engn Technol Hawthorn Vic 3122 Australia Changsha Univ Coll Elect Informat & Elect Engn Changsha 410004 Hunan Peoples R China

In this study, a multimodal diagnostic framework is developed to provide reliable early warnings of battery thermal runaway in real-world electric vehicles. Firstly, fault features of batteries are extracted from three dimensions: time, frequency and battery's internal resistance, preserving multidimensional fault information in both numerical and graphical forms. Secondly, a multimodal diagnostic algorithm for battery fault diagnosis is developed to provide early warning of thermal runaway to more than one week in advance through unsupervised learning with the joint inputs of multimodal data after graph feature extraction, dimensionality reduction and fault feature fusion. The algorithm enables the fusion and processing of fault data from different modalities, which enhances the ability to capture and analyze fault features. Thirdly, no complex pre-training or manually labeled data is required and the unified fault threshold allows for horizontal comparison. Finally, the proposed algorithm is validated using real-world data from different types of batteries in electric vehicles operating under various conditions. It shows excellent performances across different operating conditions and various battery types, achieving a balance of accuracy, reliability and generalizability.

关键词： Electric vehicles Fault diagnosis Thermal runaway early warning Multimodal fusion unsupervised learning algorithm

来源：评论

学校读者我要写书评

暂无评论

A Machine learning Approach to Estimating Queuing Delay on Routers: The Multiple-Hop Case 34

A Machine Learning Approach to Estimating Queuing Delay on R...

引用

34th International Telecommunication Networks and Applications Conference

作者： Salehin, Khondaker Ricker, Travis Wang, Yi Chen, Alex Old, Eiji Univ Portland Portland OR 97203 USA Calif State Univ Dominguez Hills Carson CA 90747 USA Kyoto Univ Kyoto Japan

ISBN: (纸本)9798350379129;9798350379112

Compression and Decompression (CoDe) is an active, state-of-the-art scheme for measuring queuing delay on Internet routers that utilizes an unsupervised learning algorithm for data processing. In this paper, we present a comparative evaluation of CoDe over a multiple-hop path under demanding traffic conditions using ns-3 simulation. This is the first evaluation of this kind because prior study of the scheme only considered a single-hop path, reflecting a limited performance evaluation. Our simulation data shows that CoDe measures queuing delay on intermediate routers along a multiple-hop path with high accuracy. It also outlines some measurement challenges prevalent in other existing schemes.

关键词： Internet performance queuing delay QoS provisioning unsupervised learning algorithm simulation

来源：评论

学校读者我要写书评

暂无评论

Exploring the heterogeneities in vehicle-involved traffic violations at intersections using latent class clustering and partial proportional odds models

引用

JOURNAL OF TRANSPORTATION SAFETY & SECURITY 2024年第1期16卷 73-95页

作者： Sun, Ya Lu, Jianfeng Ren, Gang Ma, Jingfeng Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing Peoples R China Southeast Univ Sch Transportat Nanjing Peoples R China Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China

Vehicle-involved road crashes, mainly caused by traffic violations, result in major property damage and life losses annually. Thus, it is of vital importance to clarify the heterogeneities of various factors in traffic violation severities. This study quantifies the heterogeneities across different clusters. It selects 15 potential factors from six perspectives based on 40,161 police-reported traffic violations at intersections. An unsupervised learning algorithm combining latent class clustering analysis and partial proportional odds model, as well as marginal effects, is applied to clarify the differences. Four clusters are identified as the optimal based on the Akaike information criterion, Bayesian information criterion (BIC), sample-sized adjusted BIC, and entropy-based approach. The results evidenced that the top five influential factors in traffic violations at intersections were legs of entrance, driver age, road type, area type, and signal control. Their maximum absolute values of their marginal effects were more than 35%. Furthermore, there were great heterogeneities across the clusters, especially between cluster 2 and cluster 3. Findings could provide some insightful information to prioritize effective countermeasures to mitigate the losses of traffic violations and improve road safety.

关键词： traffic violation severity heterogeneity contributing factors latent class clustering partial proportional odds model unsupervised learning algorithm

来源：评论

学校读者我要写书评

暂无评论

A lightweight algorithm for pedestrian detection in overhead images

引用

MULTIMEDIA SYSTEMS 2024年第3期30卷 158-158页

作者： Liao, Gengwei Jin, Cheng-Jie Yao, Xuejian Southeast Univ Jiangsu Key Lab Urban ITS Nanjing 210096 Jiangsu Peoples R China Jiangsu Prov Collaborat Innovat Ctr Modern Urban T Nanjing 210096 Jiangsu Peoples R China

In the field of pedestrian detection in overhead scenes, widely-used deep learning algorithms often underperform due to insufficient pedestrian feature information. To address this issue, we propose an efficient algorithm called unsupervised Overhead Dark Polar (UODP). This algorithm is specifically designed for scenes with a near 90-degree overhead angle, low pedestrian pixel resolution leading to significant loss of semantic information. It predicts the pedestrian positions by extracting prominent intensity peaks in grayscale image and utilizes polar coordinates to analyze pedestrian bounding boxes based on the top-down view characteristics. Additionally, its prominent advantage lies in unsupervised learning, which allows adaptation to various overhead scenes by adjusting intuitive parameters without training. In terms of performance, UODP has been evaluated on diverse datasets, including our open-source Southeast University overhead pedestrain dataset and several representative public datasets. Compared to cutting-edge deep learning algorithms such as DINO, RTMDet and DAB-DETR, UODP demonstrates higher accuracy and speed with lower computational requirements. The sensitivity analysis of some important parameters further reveals the mechanisms of the algorithm. Therefore, we recommend using UODP for pedestrian detection tasks that align with its design scenarios.

关键词： Pedestrian detection unsupervised learning algorithm Overhead images

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：