检索结果-内蒙古大学图书馆

Image Processing, computer Vision and Machine Learning (ICICML), International Conference on

作者： Yafei Fan Jun Liu Jian Wang Kecheng Zhou Jun Wang School of Computer Science and Technology Shenyang University of Chemical Technology Key Laboratory of Industrial Intelligence Technology on Chemical Process Shenyang China Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang China Chinese Academy of Sciences Shenyang Institute of Automation Shenyang China

ISBN: (数字)9798350355413

ISBN: (纸本)9798350355420

In terms of surface defect detection of glazed tiles, there are problems such as false detection and insufficient ability to identify small target defects due to the rich texture on the surface of glazed tiles. In response to these challenges, this article has compiled a dataset of glazed ceramic tiles and proposed an improved algorithm ECANet-FPG based on Faster R-CNN. Due to the similarity in texture and defects on the surface of ceramic tiles, this algorithm introduces the Efficient Channel Attention (ECA) module into the feature extraction network of Faster R-CNN to enhance the ability to extract deep features. At the same time, a Feature Pyramid Grids (FPG) structure is adopted to achieve efficient fusion of multi-scale features, in order to improve the recognition accuracy of small target defects. The experimental results show that the improved algorithm achieves higher detection accuracy and efficiency than the baseline model, demonstrating its effectiveness and practicality in detecting surface defects on glazed ceramic tiles.

关键词： Accuracy Target recognition Tiles Object detection Feature extraction Robustness Ceramics Surface texture Surface treatment Defect detection

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Defense Framework for Sequential Fdias in Grid-Connected Battery Energy Storage System

SSRN

引用

SSRN 2024年

作者： Xu, Shuang Li, Yuancheng Wang, Qingle Guo, Yiguo Yang, Hang School of Control and Computer Engineering North China Electric Power University Beijing102206 China Economic&Technology Research Institute State Grid Shandong Electric Power Company Shandong250022 China

With the increasing integration of battery energy storage systems (BESSs) into the power grid, BESSs are facing growing network threats, particularly sequential false data injection attacks (FDIAs). These attacks can manipulate BESS measurement data undetected, the accuracy of state of charge (SOC) estimation. To enhance the security and stability of grid-connected BESS, this paper introduces a data-driven defense framework comprising attack detection and data recovery. For the detection of sequential FDIAs, we employ the GPformer algorithm, a method that efficiently extracts temporal and spatial features from measurement data using two enhanced Transformer encoders and a gating mechanism. Following attack detection, the TACN algorithm is utilized for data recovery. This approach combines the temporal convolutional network (TCN) with an attention mechanism to predict future measurement data. By replacing the compromised data with these predictions, the TACN algorithm effectively restores the system to its normal operating state. Simulation experiments conducted on an IEEE 14-bus system integrated with BESS demonstrate the effectiveness of the framework. The results indicate that this framework not only reduces training time but also accurately detects sequential FDIAs of various intensities. Furthermore, it facilitates the recovery of attacked data, significantly decreasing the SOC estimation error of the BESS, thus providing enhanced safety for both BESSs and the power grid. © 2024, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Abstracted Model Reduction: A General Framework for Efficient Interconnected System Reduction

arXiv

引用

arXiv 2024年

作者： Poort, Luuk Janssen, Lars A.L. Besselink, Bart Fey, Rob H.B. van de Wouw, Nathan Department of Mechanical Engineering in the Dynamics & Control group Eindhoven University of Technology Eindhoven5600 MB Netherlands Bernoulli Institute for Mathematics Computer Science and Artificial Intelligence University of Groningen Groningen9700 AK Netherlands

This paper introduces the concept of abstracted model reduction: a framework to improve the tractability of structure-preserving methods for the complexity reduction of interconnected system models. To effectively reduce high-order, interconnected models, it is usually not sufficient to consider the subsystems separately. Instead, structure-preserving reduction methods should be employed, which consider the interconnected dynamics to select which subsystem dynamics to retain in reduction. However, structure-preserving methods are often not computationally tractable. To overcome this issue, we propose to connect each subsystem model to a low-order abstraction of its environment to reduce it both effectively and efficiently. By means of a high-fidelity structural-dynamics model from the lithography industry, we show, on the one hand, significantly increased accuracy with respect to standard subsystem reduction and, on the other hand, similar accuracy to direct application of expensive structure-preserving methods, while significantly reducing computational cost. Furthermore, we formulate a systematic approach to automatically determine sufficient abstraction and reduction orders to preserve stability and guarantee a given frequency-dependent error specification. We apply this approach to the lithography equipment use case and show that the environment model can indeed be reduced by over 80% without significant loss in the accuracy of the reduced interconnected model. © 2024, CC BY-NC-ND.

关键词： Structural dynamics

来源：评论

学校读者我要写书评

暂无评论

MANEUVER DECISION-MAKING WITH TRAJECTORY STREAMS PREDICTION FOR AUTONOMOUS VEHICLES

arXiv

引用

arXiv 2024年

作者： Jamal, Mais Panov, Aleksandr Cognitive Modeling Center Moscow Institute of Physics and Technology Dolgoprudny141701 Russia Federal Research Center "Computer Science and Control" Russian Academy of Sciences Moscow119333 Russia AIRI Moscow123112 Russia

Decision-making, motion planning, and trajectory prediction are crucial in autonomous driving systems. By accurately forecasting the movements of other road users, the decision-making capabilities of the autonomous system can be enhanced, making it more effective in responding to dynamic and unpredictable environments and more adaptive to diverse road scenarios. This paper presents the FFStreams++ approach for decision-making and motion planning of different maneuvers, including unprotected left turn, overtaking, and keep-lane. FFStreams++ is a combination of sampling-based and search-based approaches, where iteratively new sampled trajectories for different maneuvers are generated and optimized, and afterward, a heuristic search planner is called, searching for an optimal plan. We model the autonomous diving system in the Planning Domain Definition Language (PDDL) and search for the optimal plan using a heuristic Fast-Forward planner. In this approach, the initial state of the problem is modified iteratively through streams, which will generate maneuver-specific trajectory candidates, increasing the iterating level until an optimal plan is found. FFStreams++ integrates a query-connected network model for predicting possible future trajectories for each surrounding obstacle along with their probabilities. The proposed approach was tested on the CommonRoad simulation framework. We use a collection of randomly generated driving scenarios for overtaking and unprotected left turns at intersections to evaluate the FFStreams++ planner. The test results confirmed that the proposed approach can effectively execute various maneuvers to ensure safety and reduce the risk of collisions with nearby traffic agents. © 2024, CC BY.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

A Data-Driven Defense Model Against Sequential Fdias in Grid-Connected Battery Energy Storage Systems

SSRN

引用

SSRN 2024年

The state of charge (SoC) plays a pivotal role in ensuring the stable operation of battery energy storage systems (BESSs). However, with the escalating number of BESSs integrated into the power grid, a growing apprehension arises of potential cyber threats, which may endanger the operation of BESSs. Sequential false data injection attacks (FDIAs) can manipulate the result of SoC estimation by tampering with BESS measurement data. Notably, these attacks can elude conventional bad data detection (BDD) mechanisms. To address this issue, this paper proposes a data-driven defense model, comprising two integral submodules: detection and data recovery. The detection module employs the GPformer algorithm to detect sequential FDIAs. Specifically, the GPformer algorithm employs two Transformer encoders to extract data features, coupled with the ProbSparse self-attention mechanism to further enhance the computational efficiency of Transformer encoders. Concurrently, the data recovery module uses the TACN algorithm, which is a combination of temporal convolutional networks (TCN) and attention mechanism, to restore tampered data to its original values. Simulation experiments conducted on the IEEE 14-bus system with integrated BESSs illustrate that the proposed model exhibits a notable capability to detect sequential FDIAs. Furthermore, it demonstrates an almost flawless restoration of tampered measurement data. In summation, the results suggest the potential of this data-driven model to enhance the security and reliability of BESSs operating in a grid-connected mode. © 2024, The Authors. All rights reserved.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Forget less,count better:a domain-incremental self-distillation learning benchmark for lifelong crowd counting

引用

Frontiers of Information technology & Electronic Engineering 2023年第2期24卷 187-202页

作者： Jiaqi GAO Jingqi LI Hongming SHAN Yanyun QU James ZWANG Fei-Yue WANG Junping ZHANG Shanghai Key Laboratory of Intelligent Information Processing School of Computer ScienceFudan UniversityShanghai 200433China Institute of Science and Technology for Brain-inspired Intelligence Fudan UniversityShanghai 200433China Shanghai Center for Brain Science and Brain-inspired Technology Shanghai 201210China School of Information Science and Technology Xiamen UniversityXiamen 361005China College of Information Sciences and Technology the Pennsylvania State UniversityUniversity ParkPA 16802USA State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China

Crowd counting has important applications in public safety and pandemic control.A robust and practical crowd counting system has to be capable of continuously learning with the newly incoming domain data in real-world scenarios instead of fitting one domain ***-the-shelf methods have some drawbacks when handling multiple domains:(1)the models will achieve limited performance(even drop dramatically)among old domains after training images from new domains due to the discrepancies in intrinsic data distributions from various domains,which is called catastrophic forgetting;(2)the well-trained model in a specific domain achieves imperfect performance among other unseen domains because of domain shift;(3)it leads to linearly increasing storage overhead,either mixing all the data for training or simply training dozens of separate models for different domains when new ones are *** overcome these issues,we investigate a new crowd counting task in incremental domain training setting called lifelong crowd *** goal is to alleviate catastrophic forgetting and improve the generalization ability using a single model updated by the incremental ***,we propose a self-distillation learning framework as a benchmark(forget less,count better,or FLCB)for lifelong crowd counting,which helps the model leverage previous meaningful knowledge in a sustainable manner for better crowd counting to mitigate the forgetting when new data arrive.A new quantitative metric,normalized Backward Transfer(nBwT),is developed to evaluate the forgetting degree of the model in the lifelong learning *** experimental results demonstrate the superiority of our proposed benchmark in achieving a low catastrophic forgetting degree and strong generalization ability.

关键词： Crowd counting Knowledge distillation Lifelong learning

来源：评论

学校读者我要写书评

暂无评论

Development of a novel visual servoing probe test method for fault diagnosis of printed circuit boards

引用

IFAC-PapersOnLine 2023年第2期56卷 5248-5254页

作者： B. Tipary Á. Juniki F.G. Erdős E. Takács K. Németh Centre of Excellence in Production Informatics and Control Institute for Computer Science and Control Eötvös Loránd Research Network Budapest Hungary Department of Manufacturing Science and Engineering Faculty of Mechanical Engineering Budapest University of Technology and Economics Budapest Hungary University of Pannonia Veszprém Hungary

Printed circuit board (PCB) measurement and repair is a challenging task that requires experience and expertise to perform. PCB diagnosis and repair shops employ skilled operators to carry out the corresponding measurement tasks using measuring instruments (e.g., oscilloscopes, multimeters) in order to uncover the condition of a particular product. However, these tasks are often repetitive and meticulous, and additionally, the results need to be collected and carefully documented so that the gathered experience regarding the product can be re-used when the next product of the same type arrives into the shop. Nevertheless, the diagnosis of used PCBs is less researched and current flexible automation possibilities are limited. In this paper, a novel visual servoing probe test method and measurement tool are proposed to provide a flexible solution for PCB diagnosis with a higher level of automation. The aim of the approach is to reduce the burden on the operators by carrying out the repetitive measurement tasks and automatically storing the results while leaving the responsibility of measurement profile setup to the human expert. The proposed visual servo system uses manually teached-in measurement points, where template patterns are recorded using cameras, and it is capable of compensating positioning errors in the range of a couple of millimeters. The proof of concept of the proposed method is presented through motherboard measuring experiments, with a 99.7% success rate.

关键词： Flexible and reconfigurable manufacturing systems Quality assurance and maintenance Visual servo control Printed circuit board measurement Fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

Data-driven machine learning approaches for simultaneous prediction of peak particle velocity and frequency induced by rock blasting in mining

Rock Mechanics Bulletin

引用

Rock Mechanics Bulletin 2025年第1期4卷 29-52页

作者： Yewuhalashet Fissha Prashanth Ragam Hajime Ikeda N.Kushal Kumar Tsuyoshi Adachi P.S.Paul Youhei Kawamura Department of Geosciences Geotechnologyand Materials Engineering for ResourcesGraduate School of International Resource SciencesAkita UniversityAkita010-8502Japan Department of Mining Engineering Aksum UniversityAksum7080TigrayEthiopia School of Computer Science and Engineering VIT-AP UniversityAmaravatiAndhra PradeshIndia Department of Systems Controland Information EngineeringNational Institute of TechnologyAsahikawa College2-2-1-6 Syunkodai Asahikawa city Hokkaido071-8142Japan Department of Mining Engineering Indian Institute of Technology(Indian School of Mines)Dhanbad826004India Faculty of Engineering Hokkaido UniversityKita 8Nishi 5Kita-KuSapporo0608628Japan

The vibrations generated by rock blasting are a serious and hazardous outcome of these activities,causing harmful effects on the surrounding environment as well as the nearby *** the local ecology and human communities suffer from the consequences of these *** the severity of blasting vibrations necessitates a thorough evaluation of Peak Particle Velocity(PPV)and frequency,which are essential parameters for measuring vibration *** prediction of vibration occurrence is critically ***,this study employs five machine learning models for predicting the PPV and frequency resulting from quarry *** work compares five machine learning models(XGBoost,Catboost,Bagging,Gradient Boosting,and Random Forest Regression)to choose the most efficient performance *** performance evaluation of each five machine learning models demonstrates each model achieved a performance of more than 0.90 during the testing phase,there was a strong correlation observed between the actual and the predicted *** analysis of performance metrics shows Catboost regression model demonstrate better performance prediction comparing with the other models with R^(2)=0.983,MSE=0.000078,RMSE=0.008,NRMSE=0.019,MAD=0.004,MAPE=35.197 in the PPV prediction,and R^(2)=0.975,MSE=0.000243,RMSE=0.015,NRMSE=0.031,MAD=0.008,MAPE=37.281 for the frequency *** study will help mining engineers and blasting experts to select the best machine learning model and its hyperparameters in estimating ground vibration,and *** the context of the mining and civil industry,the application of this study offers significant potential for enhancing safety protocols and optimizing operational *** employing machine learning models,this research aims to accurately predict and assess ground vibrations with frequency resulting from rock blasting.

关键词： PPV Frequency Machine learning Blasting Regression Catboost XGboost

来源：评论

学校读者我要写书评

暂无评论

Retinal disorder diagnosis based on hybrid deep learning models

引用

Multimedia Tools and Applications 2025年 1-25页

作者： Sedik, Ahmed El-Shafai, Walid El-Hag, Noha A. El-Banby, Ghada M. Abd El-Samie, Fathi E. Smart Systems Engineering Laboratory College of Engineering Prince Sultan University Riyadh11586 Saudi Arabia Department of the Robotics and Intelligent Machines Faculty of Artificial Intelligence Kafrelsheikh University Kafr el-Sheikh Egypt Computer Science Department Prince Sultan University Riyadh11586 Saudi Arabia Department of Electronics and Electrical Communications Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt The Higher Institute of Commercial Sciences Algarbia Al Mahalla Al Kubra31951 Egypt Department of Industrial Electronics and Control Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt Department of Information Technology College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia

Anomaly detection from medical images is badly needed for automated diagnosis. For example, medical images obtained with several modalities, such as magnetic resonance (MR) and confocal microscopy, need to be classified for anomaly detection. The MR images may contain tumors that need to be detected with anomaly detection techniques. In MR imaging, brain tumors are represented as areas composed of abnormal cells that need to be detected. On the other hand, confocal microscopy can be used for retinal imaging. Hence, there is a need to apply classification techniques for diagnosing diseases of the eye, such as Diabetic Retinopathy (DR), which leads to exudates in different regions of the retina. The detection of exudates is necessary for the diagnosis of DR. In this paper, two models are presented for the classification of MR and retinal images. Much research has been done in this area, but it did not achieve high accuracy of classification in an acceptable time. In addition, the previous models did not consider both types of image modalities considered in this paper. The suggested models are Convolutional Neural Network (CNN) and Convolutional Long Short-Term Memory (Conv-LSTM). They are used to classify normal and abnormal cases of both brain tumor and DR diseases. The CNN model is composed of six convolutional layers and six max-pooling layers. A soft-max activation function takes the classification decision. The proposed Conv-LSTM model is composed of three stages: pre-processing, feature extraction, and classification. The deep learning model performance is better than those of traditional networks. The CNN gives an accuracy of 94.26% for DR classification and 95.8% for brain tumor classification, while the Conv-LSTM gives the maximum accuracy of 96.41% for DR classification and 97.4% for brain tumor classification. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Confocal microscopy

来源：评论

学校读者我要写书评

暂无评论

Multi Modal EEG-Based Classification of Alzheimer's Disease and Mild Cognitive Impairment Using Hilbert Huang Transform

Multi Modal EEG-Based Classification of Alzheimer's Disease ...

引用

Advances in Modern Age Technologies for Health and Engineering Science (AMATHE), International Conference on

作者： G. Gopu S. Sathish T. S. Shanmathan G. Suthakar N. Sivakumaran M S Geetha Devasena Electronics and Communication Engineering Sri Ramakrishna Engineering College Coimbatore India Instrumentation and Control Engineering National Institute of Technology Tiruchirappalli India Computer Science and Engineering Sri Ramakrishna Engineering College Coimbatore India

ISBN: (数字)9798350371567

ISBN: (纸本)9798350371574

This study investigates whether high-resolution brain-functional connectivity, which is measured by using Electroencephalography (EEG) combined with Hilbert Huang Transform (HHT) can be useful in Alzheimer Disease Diagnosis. The major issue with Alzheimer's disease (AD) is that it leads to cognitive decline and affects millions of people worldwide. Early prediction of the disease is mandatory and crucial for timely intervention and effective care. EEG signals, a non-invasive and cost-effective method to assess brain activity, have shown potential in predicting this disease. In this study, we propose an approach that utilizes EEG data combined with Hilbert-Huang Transform (HHT) and machine learning algorithms for classification which is a novelty of the study that enables early prediction of Alzheimer's disease. Our methodology was tested on a dataset comprising EEG signals recorded using 19 electrodes from both healthy individuals and patients with Alzheimer's and the Mild Cognitive Impairment (MCI). The results were promising, demonstrating high accuracy, sensitivity, and specificity. Machine learning techniques can be employed for categorizing EEG signals, but it essential to validate their effectiveness using large and diverse datasets. This research highlights the potential of HHT and EEG data in the prediction of this disease. Principal Component Analysis (PCA) is used for dimensional reduction of the dataset. Notably EEG signals combined with Hilbert Huang Transform and Principal Component Analysis gives better accuracy of 98.6% for KNN. This research highlights the potential of EEG analysis combined with advanced techniques for accurate classification of dementia-related disorders.

关键词： Accuracy Machine learning algorithms Empirical mode decomposition Transforms Machine learning Sensitivity and specificity Electroencephalography

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：