检索结果-内蒙古大学图书馆

8th international conference on Electronic Information Technology and Computer engineering, EITCE 2024

作者： Fan, Mingxing Lin, Kairong Lin, Linhan Wu, Longhui Feng, Hao Jimei University Fujian Xiamen China

ISBN: (纸本)9798400710094

During the production process, inkjet characters on product packaging often exhibit defects, such as missing or smudged characters, due to mechanical vibrations and ink quality issues. Therefore, it is necessary to conduct inkjet detection before products leave the factory. Traditional manual inspection methods are inefficient, costly, and prone to fatigue, making them unsuitable for long-term tasks. Moreover, conventional machine vision technologies cannot meet the required detection accuracy for inkjet characters in complex backgrounds. To address the imbalance between speed and accuracy in text detection algorithms under complex conditions, this paper presents a detailed study on inkjet character detection and proposes a lightweight DBNet-based segmentation algorithm that significantly improves detection speed and accuracy. © 2024 Copyright held by the owner/author(s).

关键词： machine vibrations

来源：评论

学校读者我要写书评

暂无评论

MTMDC-GAN: Self-Attention Driven Multi-Scale Temporal Synthesis with Multi-Domain Analysis and Contrastive learning

MTMDC-GAN: Self-Attention Driven Multi-Scale Temporal Synthe...

引用

2025 IEEE international conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhi, Weihai He, Kejing School of Computer Science and Engineering South China University of Technology Guangzhou China

ISBN: (纸本)9798350368741

The synthesis of high-quality multivariate time series (MTS) is critical for enhancing the performance of predictive models and data-driven decision-making. This paper introduces a novel approach to MTS generation, integrating multi-scale data fusion with self-attention mechanisms to capture complex interdependencies across temporal scales, which to some extent aids the generator in producing coherent MTS, improving the representational capacity of the data. The methodology extends beyond conventional spatial feature extraction by incorporating Fourier Transformation analysis in the frequency domain and Contrastive learning in the time domain. This multi-domain approach uncovers deeper insights into the underlying patterns and cyclic behaviors of time series data. Extensive experimentation across different datasets and configurations demonstrate superiority over existing techniques, achieving greater accuracy and realism in synthesized MTS. © 2025 IEEE.

关键词： Contrastive learning Fourier Transformation Multi-Scale Synthesis

来源：评论

学校读者我要写书评

暂无评论

Detection of Alzheimer’s disease and Frontotemporal Dementia: An Explainable machine learning Approach Using EEG signals 13

Detection of Alzheimer’s disease and Frontotemporal Dementi...

引用

13th international Winter conference on Brain-Computer Interface, BCI 2025

作者： Lee, Hwa-Yeon Jung, Min-Kyung Lee, Choel-Hui Kim, Hakseung Kim, Dong-Joo Department of Brain and Cognitive Engineering Interdisciplinary Program in Precision Public Health Korea University Seoul Korea Republic of Department of Brain and Cognitive Engineering Korea University Seoul Korea Republic of

ISBN: (纸本)9798331521929

Alzheimer’s disease and frontotemporal dementia are among the most prevalent neurodegenerative diseases worldwide. With the absence of definitive treatments, early detection and precise diagnosis remain crucial for alleviating patient suffering and reducing healthcare costs. This study aimed to differentiate Alzheimer’s disease, frontotemporal dementia and normal cognition using electroencephalography biomarkers. Significant power spectral density and functional connectivity features were identified via one-way analysis of variance and Scheffe’s post-hoc test. A three-class classification was performed by a support vector machine classifier and explainable artificial intelligence technique was employed to investigate feature importance. This classifier achieved an average accuracy of 84.09%, and linear shapley additive explanation model identified T5-Fp1 connectivity in alpha band as the most important feature, consistent with findings from previous studies. Further clinical research and brain network analysis using advanced deep learning models could enhance our understanding of the cognitive functions and neural mechanisms underlying these dementias. © 2025 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Pavement Crack Segmentation using Incomplete Supervision Network 24

Pavement Crack Segmentation using Incomplete Supervision Net...

引用

3rd international conference on Artificial Intelligence and Intelligent Information Processing, AIIIP 2024

作者： Wu, Wenxiu Jin, Yihui Fang, Zhihua Zhang, Jianfeng Fan, Xiaqi Song, Xiaoliang Port and Transportation Management Center Zhejiang Jinhua China Lanxi Transportation Construction Investment Group Co. Ltd. Zhejiang Jinhua China Lanxi Tongda Highway Maintenance Engineering Co. Ltd. Zhejiang Jinhua China Lanxi SHUNDA Road and Bridge Engineering Co. Ltd. Zhejiang Jinhua China

ISBN: (纸本)9798400707308

Challenges such as the lack of baseline pavement disease databases and limited segmentation accuracy and speed impact the performance of pavement disease segmentation and its practical engineering applications. To explore the feasibility of incomplete supervised learning, an experimental study generated inaccurate labels in batches by combining traditional image segmentation algorithms with manual methods, and the Mean Teacher model was tested. For imprecise supervised learning, the test F1 accuracy reached 0.572 after adding imprecise labels to the training set of self-collected data, although external imprecise labels were not feasible. In incomplete supervised learning, the Mean Teacher model showed significant advantages, achieving a test accuracy (MIoU) of 0.796 and an F1 score of 0.596 on self-collected data. Using unlabeled external images can enhance network accuracy in semantic segmentation tasks. With a small number of accurately labeled images, incomplete supervised learning can achieve substantial prediction accuracy. © 2024 Copyright held by the owner/author(s).

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Medical BCFL: A blockchain-enabled federated learning architecture for secure healthcare data sharing 24

Medical BCFL: A blockchain-enabled federated learning archit...

引用

2nd international conference on Artificial Intelligence, Systems and Network Security, AISNS 2024

作者： Dong, Qianyu Liu, Jiazheng Jiang, Yuchen School of Cyber Science and Technology Shenzhen Campus Sun Yat-sen University Guangdong Shenzhen China School of Computer Science and Technology DongHua University Shanghai China

ISBN: (纸本)9798400711237

In response to the growing demand for secure healthcare systems, this paper disaplys a blockchain-enabled federated learning architecture called Medical BCFL. The architecture aims to address privacy concerns and facilitate secure healthcare data sharing by combining the decentralized nature of blockchain with the collaborative learning model of federated learning. First, a blockchain system is built through smart contracts to record model updates and participant reputations for each round of communication. Next, a federated learning system is built to ensure privacy protection during collaborative model training and encourage data sharing through a reward mechanism. Ultimately, a loosely coupled architecture is used to integrate the blockchain with the federated learning system, where model training takes place in the federated learning system and intermediate information is synchronized to the blockchain. The proposed method achieves over 92% accuracy in MNIST dataset in an adversarial setting of 40% malicious clients and 8% accuracy loss in a real-world BloodMNIST dataset with the same proportion of malicious clients. © 2024 Copyright held by the owner/author(s).

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A Methodology for Automated Conversion of Axis-Aligned to Polygonal and Oriented Bounding Box Annotations in Aerial Imagery Object Detection 25th

A Methodology for Automated Conversion of Axis-Aligned to Po...

引用

25th international conference on Intelligent data engineering and Automated learning

作者： Freire, Daniela L. de Carvalho, Andre C. P. L. F. Peterlevitz, Augusto Jose Chinelatto, Mateus Antonio da Silva, Ricardo Dutra Rojas Perea, Juan Fernando Univ Sao Paulo Inst Math & Comp Sci Sao Carlos SP Brazil Inst Pesquisas Eldorado Campinas Brazil Transmissora Alianca Energia Elet SA Belo Horizonte MG Brazil

ISBN: (纸本)9783031777370;9783031777387

Object detection in aerial imagery presents significant challenges in computer vision due to the varied orientations and complex backgrounds of objects such as buildings and vehicles. Current annotation tools often fail to accurately delineate these objects, relying on manual bounding box methods that are both time-consuming and inconsistent. Our novel methodology automates the conversion of axis-aligned annotations into polygonal and rotated annotations, prioritising systematic and scalable enhancements to data quality rather than modifying the model itself. Precise annotations, crucial for determining object locations and boundaries, are fundamental to this approach. We evaluated this methodology through a case study involving electrical transmission towers in aerial images, using advanced object detectors based on variations of the YOLOv8 algorithm. Preliminary results indicate that our automated method not only improves annotation accuracy but also significantly reduces the manual effort required, thereby lowering overall costs and time for data preparation in object detection training. The success of this methodology underscores its potential for broader applications and further advancements in automated annotation technologies.

关键词： Object Detection Image Segmentation machine learning YOLO Model Segment Anything Model Oriented Bounding Box Polygonal Annotation Annotation Conversion

来源：评论

学校读者我要写书评

暂无评论

Flotation froth content estimation on the basis of a machine learning process 23

Flotation froth content estimation on the basis of a Machine...

引用

23rd Slovak-Czech-Polish Optical conference on Wave and Quantum Aspects of Contemporary Optics

作者： Galas, Jacek Litwin, Dariusz Kozielski, Michal Foszcz, Dariusz Saramak, Daniel Lenczowski, Stanislaw Kasińska-Pilut, Ewelina Prętki, Jaroslaw Lukasiewicz Research Network Tele and Radio Research Institute 11 Ratuszowa St. Warsaw03-450 Poland Lukasiewicz Research Network Institute of Innovative Technologies EMAG Poland AGH University of Cracow 30 Mickiewicza Ave. Cracow30-059 Poland KGHM Polska Miedź S.A. Division of Concentrators 1 Kopalniana St Polkowice59-100 Poland

ISBN: (数字)9781510688094

ISBN: (纸本)9781510688087

The flotation froth content assessment can be carried out in various ways. One of them is the machine learning (ML) process applied to the preprocessed and parametrized flotation froth images. The ML procedure can be conducted on the basis of Artificial Neural Network (ANN) or Linear Discriminant Analysis (LDA). This paper presents the LDA application in a ML process. The ML algorithms have their origins in the family of classification algorithms. They are constructed on the basis of training groups of data and enable us to classify the unknown data to one of the training groups. Theoretically, if we have more training groups we should be able to obtain more accurate estimation of data. However, such approach is not efficient. Each training group should have a high number of data sets, which in most cases is difficult to obtain. This paper presents the estimation of the flotation froth content in the mineral processing plant on the basis of images of the froth surface. The experiment was performed in the Pb Mineral Processing Plant. The images of the flotation froth surface were registered for seven different configurations of the flotation process parameters. The flotation parameters were stabilized 30 minutes before the image registration process. The images registered in the stabilized technological conditions constituted the training group of images for the ML process. For each of the technological configurations of the flotation process, 10 collective froth samples were collected for direct chemical analysis. This means that each of 7 training groups of the froth images contained 10 subgroups of images with well-defined froth content. The ML process enabled us to construct the froth content estimation algorithms for the flotation process in the experiment. © 2025 SPIE.

关键词： Image registration

来源：评论

学校读者我要写书评

暂无评论

Graph-theoretical investigation of trajectory dynamics and size characteristics in tropical cyclones

引用

Natural Hazards 2025年 1-18页

作者： Wang, Yixiang Wang, Jiayao Chang, Yu Cai, Kang Li, Sunwei Dong, You Department of Civil and Environmental Engineering The Hong Kong University of Science and Technology Kowloon Clear Water Bay Hong Kong Department of Civil and Environmental Engineering The Hong Kong Polytechnic University Kowloon Hung Hom Hong Kong Institute for Ocean Engineering Tsinghua Shenzhen International Graduate School Guangdong Shenzhen China Institute of Structural Engineering College of Civil Engineering and Architecture Zhejiang University Hangzhou China School of Civil Engineering and Architecture Guangxi University Nanning China

The intensification of climate changes has led to increased tropical cyclone (TC) intensities and subsequent damage, emphasizing the critical need for accurate trajectory prediction to mitigate their impact. In this study, a graph-theory-based approach was employed for the identification of TC trajectory. Using reanalysis data, each targeted TC can be constructed as a graph during its TC lifetime. Four graph metrics are computed from each graph constructed using different data sources, including mean sea level pressure, wind speed, and total precipitation. Among the graphs constructed, those representing mean sea level pressure (MSLP) and wind speed at 10 m (WD10) graphs show superior advantages in identifying TC trajectory. Furthermore, the metric PageRank of MSLP graph even reveals a notable ability to estimate TC size. Comparisons with a similar graph-theoretical approach demonstrate that our method exhibits superior performance in capturing complex TC dynamics. We anticipate to integrating the graph-theory-based approach into machine learning models to enhance the accuracy of predicting TC trajectories and intensities in future studies. © The Author(s) 2025.

关键词： Forecast model machine learning Trajectory Tropical cyclone

来源：评论

学校读者我要写书评

暂无评论

Simulation and implementation of deep learning based expression recognition

Simulation and implementation of deep learning based express...

引用

2024 international conference on Mechatronic engineering and Artificial Intelligence, MEAI 2024

作者： Qin, Lijuan Wang, Shiyi Chen, Jiayu School of Information Science and Engineering Shenyang Ligong University Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Science Shenyang China

ISBN: (纸本)9781510689176

Facial expressions are a challenging area in image recognition, and ensuring the universality of the model in the presence of a large sample size is a difficult point in such research. This article conducts repeatability experiments on two extreme scenarios: high-quality small samples and low-quality large samples, to explore solutions. Through transfer learning, data augmentation, and substantive evaluation of confusion matrices, the article demonstrates the accurate classification of seven facial expressions and the design of a 4-class model with high universality, which is presented in the form of an APP. © 2025 SPIE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An Investigation into machine learning-Based Techniques for Detecting Malicious Cryptomining Activities 3rd

An Investigation into Machine Learning-Based Techniques for ...

引用

3rd international conference on machine learning, Cloud Computing and Intelligent Mining, MLCCIM 2024

作者： Xie, Renchao Huang, Xuange Tang, Qinqin Jiangsu Future Network Innovation Institute Nanjing China Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9789819616978

With the popularity of blockchain concepts and the appreciation of cryptocurrencies, an increasing number of hackers and illegal elements have started to focus on stealing users’ hardware resources for mining activities. In this context, web mining has emerged, which can occupy the computing resources of website visitors to mine cryptocurrencies. However, this method can affect user experience, increase energy consumption, and reduce computer performance, making web mining detection and defense increasingly important. Due to the high degree of obfuscation and complex behavior in web mining code, this paper proposes a mining detection model based on machine learning, which extracts features from multiple dimensions and combines machine learning algorithms to identify web pages with mining behavior. We used four classification algorithms for model training, and through extensive experiments and performance comparison, the Random Forest algorithm was proven to be the most effective [1]. This indicates its high accuracy and suitability for detecting malicious mining activities. Our research not only provides a robust method for identifying malicious web mining but also lays a foundation for future cybersecurity research, offering insights into the application of machine learning techniques in this domain. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： blockchain cryptojacking cryptomining detection machine learning

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：