检索结果-内蒙古大学图书馆

Bloch Equation Enables Physics-informed Neural Network in Parametric Magnetic Resonance Imaging

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Cai, Qingrui Zhu, Liuhong Zhou, Jianjun Qian, Chen Guo, Di Qu, Xiaobo National Integrated Circuit Industry Education Integration Innovation Platform School of Electronic Science and Engineering National Model Microelectronics College Xiamen University Xiamen361102 China Fudan University Xiamen361015 China Department of Electronic Science Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance Xiamen University Xiamen361102 China School of Computer and Information Engineering Fujian Engineering Research Center for Medical Data Mining and Application Xiamen University of Technology Xiamen361024 China

Magnetic resonance imaging (MRI) is an important non-invasive imaging method in clinical diagnosis. Beyond the common image structures, parametric imaging can provide the intrinsic tissue property thus could be used in quantitative evaluation. The emerging deep learning approach provides fast and accurate parameter estimation but still encounters the lack of network interpretation and enough training data. Even with a large amount of training data, the mismatch between the training and target data may introduce errors. Here, we propose one way that solely relies on the target scanned data and does not need a pre-defined training database. We provide a proof-of-concept that embeds the physical rule of MRI, the Bloch equation, into the loss of physics-informed neural network (PINN). PINN enables learning the Bloch equation, estimating the T2 parameter, and generating a series of physically synthetic data. Experimental results are conducted on phantom and cardiac imaging to demonstrate its potential in quantitative MRI. Copyright © 2023, The Authors. All rights reserved.

关键词： Magnetic resonance imaging

Computing optimal partition problems via Lagrange multiplier approach

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Cheng, Qing Guo, Jing Wang, Dong Department of Mathematics Tongji University Shanghai200092 China Key Laboratory of Intelligent Computing and Applications Tongji University Ministry of Education China School of Science and Engineering The Chinese University of Hong Kong Guangdong Shenzhen518172 China Shenzhen International Center for Industrial and Applied Mathematics Shenzhen Research Institute of Big Data Guangdong518172 China

In this paper, we consider numerical approximations for the optimal partition problem using Lagrange multipliers. By rewriting it into constrained gradient flows, three and four steps numerical schemes based on the Lagrange multiplier approach [16, 17] are proposed to solve the constrained gradient system. Numerical schemes proposed for the constrained gradient flows satisfy the nice properties of orthogonality-preserving, norm-preserving, positivity-preserving and energy dissipating. The proposed schemes are very efficient in which only linear Poisson equations are solved at each time step. Extensive numerical results in 2D and 3D for optimal partition problem are presented to validate the effectiveness and accuracy of the proposed numerical schemes. Copyright © 2024, The Authors. All rights reserved.

关键词： Lagrange multipliers

Correction to: A remote sensing image encryption and compression scheme using novel hyperchaotic system and plaintext related random S-box

学校读者我要写书评

暂无评论

Nonlinear Dynamics 2024年第3期113卷 2791-2791页

作者： Li, Zhen Yang, Siqi Tan, Weijie Wu, Xianming College of Big Data and Information Engineering Guizhou University Guiyang China Key Laboratory of Advanced Manufacturing Technology of Ministry of Education State Key Laboratory of Public Big Data Guizhou University Guiyang China School of Mechanical and Electrical Engineering Guizhou Normal University Guiyang China

WiOpen: A Robust Wi-Fi-based Open-set Gesture Recognition Framework

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Xiang Huang, Jingyang Yan, Huan Zhao, Peng Zhuang, Guohang Liu, Zhi Liu, Bin CAS Key Laboratory of Electromagnetic Space Information University of Science and Technology of China Hefei230026 China School of Computer and Information Hefei University of Technology Hefei230601 China School of Big Data and Computer Science Guizhou Normal University Guiyang550025 China Department of Computer and Network Engineering The University of Electro-Communications Tokyo1828585 Japan

Recent years have witnessed a growing interest in Wi-Fi-based gesture recognition. However, existing works have predominantly focused on closed-set paradigms, where all testing gestures are predefined during training. This poses a significant challenge in real-world applications, as unseen gestures might be misclassified as known classes during testing. To address this issue, we propose WiOpen, a robust Wi-Fi-based Open-Set Gesture Recognition (OSGR) framework. Implementing OSGR requires addressing challenges caused by the unique uncertainty in Wi-Fi sensing. This uncertainty, resulting from noise and domains, leads to widely scattered and irregular data distributions in collected Wi-Fi sensing data. Consequently, data ambiguity between classes and challenges in defining appropriate decision boundaries to identify unknowns arise. To tackle these challenges, WiOpen adopts a two-fold approach to eliminate uncertainty and define precise decision boundaries. Initially, it addresses uncertainty induced by noise during data preprocessing by utilizing the CSI ratio. Next, it designs the OSGR network based on an uncertainty quantification method. Throughout the learning process, this network effectively mitigates uncertainty stemming from domains. Ultimately, the network leverages relationships among samples’ neighbors to dynamically define open-set decision boundaries, successfully realizing OSGR. Comprehensive experiments on publicly accessible datasets confirm WiOpen’s effectiveness. Notably, WiOpen also demonstrates superiority in cross-domain tasks when compared to state-of-the-art approaches. © 2024, CC BY-SA.

关键词： Learning systems

Location Rating Based on Graph Neural Network 2020

学校读者我要写书评

暂无评论

Location Rating Based on Graph Neural Network

Proceedings of the 2020 International Conference on Cyberspace Innovation of Advanced Technologies

作者： Fengbin Zheng Li Ding Lanxue Dang Henan Key Laboratory of Big Data Analysis and Processing School of Computer and Information Engineering Henan University Kaifeng PR China

ISBN: (纸本)9781450387828

With the rapid development of social networks, people are increasingly interested in sharing their location. Location recommendation has become an important personalization service for location-based social networks (LBSN). Location rating is one of the important tools. However, location-based social networks contain multidimensional network structures and node information. Existing approaches mostly focus on network structures that utilize one of these dimensions, which makes it difficult to efficiently aggregate information from multiple dimensions simultaneously. To overcome these difficulties, one of the recent approaches is the social recommendation based on graph neural networks (GNNs). Based on this, this paper proposes a graph neural network framework for location rating. Graph neural networks are highly inductive and can efficiently aggregate network structure and node information. In particular, the method not only aggregates homogeneous social networks composed of user and heterogeneous bipartite graphs composed of user-location, but also constructs location networks using location sequences to propose an aggregation model for location networks. Experiments are conducted on two datasets, and the results show that the method improves on the root mean square error (RMSE) and mean absolute error (MAE) metrics.

关键词： Graph neural network Location rating Location based social networks

RNaAE: A Novel Approach for Identifying Unseen Railway Anomalies

学校读者我要写书评

暂无评论

RNaAE: A Novel Approach for Identifying Unseen Railway Anoma...

International Conference on Signal Processing Proceedings (ICSP)

作者： Jiayu Zhang Qingji Guan Junbo Liu Yaping Huang Jianyong Guo Beijing Key Laboratory of Traffic Data Analysis and Mining Beijing Jiaotong University Beijing China Infrastructure Inspector Center China Academy of Railway Science Corporation Limited Beijing China Tangshan Research Institute of Beijing Jiaotong University Tangshan China Urban Rail Transit National Engineering Research Center for Digital Construction and Evaluation Tianjin China

ISBN: (数字)9798350387384

ISBN: (纸本)9798350387391

Detecting abnormal objects in railway track inspection images using vision-based technology is a crucial task for ensuring the safety of railway transportation. Traditional supervised object detection methods fail to achieve satisfactory results due to the diverse categories of abnormal objects and the lack of abnormal samples. Even though the widely used Autoencoder can leverage reconstruction errors to detect anomalies without using abnormal data, they tend to generate a relatively high number of false positives. In this paper, we address the task in an unsupervised manner and propose a novel Random Network-Assisted Autoencoder, called RNaAE, for identifying unseen abnormal objects. Specifically, we first design a learnable network to fit a randomly initialized stochastic network with fixed weights, where the difference between two predictions can then be used to estimate whether the candidate object is anomalous. After combined with a traditional Autoencoder, a Gaussian mixture model is then used to classify the candidate box into normal and abnormal by anomaly scores. Extensive experiments conducted on our collected railway anomaly dataset demonstrate that the proposed RNaAE exceeds previous stateof-the-art methods, achieving 98.23% and 92.02% in terms of AUROC and F1-score.

关键词： Autoencoders Transportation Stochastic processes Object detection Signal processing Inspection Rail transportation Safety Object recognition Anomaly detection

Federated Unlearning with Gradient Descent and Conflict Mitigation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Pan, Zibin Wang, Zhichao Li, Chi Zheng, Kaiyan Wang, Boqi Tang, Xiaoying Zhao, Junhua The School of Science and Engineering The Chinese University of Hong Kong Shenzhen China The Cyberspace Academy Guangzhou University China The Shenzhen Institute of Artificial Intelligence and Robotics for Society China The Guangdong Provincial Key Laboratory of Future Networks of Intelligence China Department of Mathematics Department of Statistics University of Michigan United States Shenzhen Research Institute of Big Data China

Federated Learning (FL) has received much attention in recent years. However, although clients are not required to share their data in FL, the global model itself can implicitly remember clients’ local data. Therefore, it’s necessary to effectively remove the target client’s data from the FL global model to ease the risk of privacy leakage and implement "the right to be forgotten". Federated Unlearning (FU) has been considered a promising way to remove data without full retraining. But the model utility easily suffers significant reduction during unlearning due to the gradient conflicts. Furthermore, when conducting the post-training to recover the model utility, the model is prone to move back and revert what has already been unlearned. To address these issues, we propose Federated Unlearning with Orthogonal Steepest Descent (FedOSD). We first design an unlearning Cross-Entropy loss to overcome the convergence issue of the gradient ascent. A steepest descent direction for unlearning is then calculated in the condition of being non-conflicting with other clients’ gradients and closest to the target client’s gradient. This benefits to efficiently unlearn and mitigate the model utility reduction. After unlearning, we recover the model utility by maintaining the achievement of unlearning. Finally, extensive experiments in several FL scenarios verify that FedOSD outperforms the SOTA FU algorithms in terms of unlearning and model utility. © 2024, CC BY.

关键词： Federated learning

A Fast Filling Algorithm for Image Restoration Based on Contour Parity

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2020年第4期63卷 509-519页

作者： Yan Liu Wenxin Hu Longzhe Han Maksymyuk Taras Zhiyun Chen School of Data Science and Engineering East China Normal UniversityShanghai200062China JiangXi Province Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of TechnologyNanchang330099China Department of Telecommunication Institute of Telecommunication Radio Electronics and Electronic EngineeringLviv Polytechnic National UniversityLviv79013Ukraine

Filling techniques are often used in the restoration of *** the existing filling technique approaches either have high computational costs or present problems such as filling holes *** paper proposes a novel algorithm for filling holes and regions of the *** proposed algorithm combines the advantages of both the parity-check filling approach and the region-growing inpainting *** points of the region’s boundary are used to search and to fill the *** scanning range of the filling method is within the target *** proposed method does not require additional working memory or assistant colors,and it can correctly fill any complex *** results show that,compared to other approaches,the proposed algorithm fills regions faster and with lower computational cost.

关键词： Region filling image restoration parity check region growing

A prediction-correction based iterative convolution-thresholding method for topology optimization of heat transfer problems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Huangxin Dong, Piaopiao Wang, Dong Wang, Xiao-Ping School of Mathematical Sciences Fujian Provincial Key Laboratory on Mathematical Modeling and High Performance Scientific Computing Xiamen University Fujian361005 China School of Science and Engineering The Chinese University of Hong Kong Guangdong Shenzhen518172 China Shenzhen International Center for Industrial and Applied Mathematics Shenzhen Research Institute of Big Data Guangdong518172 China

In this paper, we propose an iterative convolution-thresholding method (ICTM) based on prediction-correction for solving the topology optimization problem in steady-state heat transfer equations. The problem is formulated as a constrained minimization problem of the complementary energy, incorporating a perimeter/surface-area regularization term, while satisfying a steady-state heat transfer equation. The decision variables of the optimization problem represent the domains of different materials and are represented by indicator functions. The perimeter/surface-area term of the domain is approximated using Gaussian kernel convolution with indicator functions. In each iteration, the indicator function is updated using a prediction-correction approach. The prediction step is based on the variation of the objective functional by imposing the constraints, while the correction step ensures the monotonically decreasing behavior of the objective functional. Numerical results demonstrate the efficiency and robustness of our proposed method, particularly when compared to classical approaches based on the ICTM. Copyright © 2024, The Authors. All rights reserved.

关键词： Convolution