检索结果-内蒙古大学图书馆

Hierarchical Dynamic Graph clustering Network

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第9期36卷 4722-4735页

作者： Chen, Jie Jiao, Licheng Liu, Xu Li, Lingling Liu, Fang Chen, Puhua Yang, Shuyuan Hou, Biao Xidian Univ Int Res Ctr Intelligent Percept & Computat Sch Artificial Intelligence Key Lab Intelligent Percept & Image Understanding Xian 710071 Shaanxi Provinc Peoples R China

Connections between visual components are ubiquitous. Graphs, as a highly flexible data structure, not only allow imposing relational induction bias on data, but can provide a completely distinct learning perspective for regular image data. In this article, we propose a hierarchical dynamic graph clustering network (HDGCN) for visual feature learning. We construct hierarchical graph representations in graph domain in an adaptive, data-adaptive and task-adaptive manner. First, the initial graph is constructed in high-dimensional feature domain of images. To mine the hierarchical geometric features in latent graph space, adaptive clustering network (ClusterNet) is performed to learn discriminative clusters and generates cluster-based coarse graph. Then, graph convolutional networks (GCNs) are used to diffuse, transform and aggregate information among clusters. So, the intra-class and inter-class information is fully explored to increase the discriminativity of graph representations. Next, coarsened graph representations are mapped to grid based on its affinity with linear projection features. To further improve the task adaptation of clusters and hierarchical graph representations, ClusterNet and GCNs are fused in the same framework for end-to-end training and clusters is updated dynamically. We have conducted extensive experiments on classification and segmentation tasks. The experimental results fully validate the robustness of the proposed algorithm.

关键词： Task analysis Visualization Feature extraction clustering algorithms Semantics Convolutional neural networks Convolution Graph coarsening graph convolutional networks hierarchical graph representation learning spectral clustering

来源：评论

学校读者我要写书评

暂无评论

Accurate Registration of Cross-Modality Geometry via Consistent clustering

引用

IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS 2024年第7期30卷 4055-4067页

作者： Zhao, Mingyang Huang, Xiaoshui Jiang, Jingen Mou, Luntian Yan, Dong-Ming Ma, Lei Chinese Acad Sci Beijing Acad Artificial Intelligence Inst Automat Beijing 100045 Peoples R China Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing 100045 Peoples R China Shanghai AI Lab Shanghai 200433 Peoples R China Chinese Acad Sci State Key Lab Multimodal Artificial Intelligence Inst Automat Beijing 100045 Peoples R China Univ Chinese Acad Sci Sch AI Beijing 101408 Peoples R China Beijing Univ Technol Beijing Inst Artificial Intelligence Beijing Key Lab Multimedia & Intelligent Software Beijing 100021 Peoples R China Peking Univ Natl Biomed Imaging Ctr Beijing 100871 Peoples R China Peking Univ Sch Comp Sci Beijing Acad Artificial Intelligence Beijing 100871 Peoples R China Peking Univ Sch Comp Sci Natl Key Lab Multimedia Informat Proc Beijing 100871 Peoples R China

The registration of unitary-modality geometric data has been successfully explored over past decades. However, existing approaches typically struggle to handle cross-modality data due to the intrinsic difference between different models. To address this problem, in this article, we formulate the cross-modality registration problem as a consistent clustering process. First, we study the structure similarity between different modalities based on an adaptive fuzzy shape clustering, from which a coarse alignment is successfully operated. Then, we optimize the result using fuzzy clustering consistently, in which the source and target models are formulated as clustering memberships and centroids, respectively. This optimization casts new insight into point set registration, and substantially improves the robustness against outliers. Additionally, we investigate the effect of fuzzier in fuzzy clustering on the cross-modality registration problem, from which we theoretically prove that the classical Iterative Closest Point (ICP) algorithm is a special case of our newly defined objective function. Comprehensive experiments and analysis are conducted on both synthetic and real-world cross-modality datasets. Qualitative and quantitative results demonstrate that our method outperforms state-of-the-art approaches with higher accuracy and robustness. Our code is publicly available at https://***/zikai1/CrossModReg.

关键词： Point cloud compression Geometry Three-dimensional displays Laser radar Tensors clustering algorithms Solid modeling Cross-modality geometry point cloud registration 3D reconstruction adaptive fuzzy clustering CAD

来源：评论

学校读者我要写书评

暂无评论

Multiview Fuzzy clustering Based on Anchor Graph

引用

IEEE TRANSACTIONS ON FUZZY SYSTEMS 2024年第3期32卷 755-766页

作者： Yu, Weizhong Xing, Liyin Nie, Feiping Li, Xuelong Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Sch Cybersecur Xian 710072 Peoples R China Northwestern Polytech Univ Key Lab Intelligent Interact & Applicat Minist Ind & Informat Technol Xian 710072 Peoples R China Northwestern Polytech Univ Sch Cybersecur Xian 710072 Peoples R China Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Xian 710072 Peoples R China

With the development of information technology, a large number of multiview data has emerged, which makes multiview clustering algorithms considerably attractive. Previous graph-based multiview clustering methods usually contain two steps: obtaining the fusion graph or spectral embedding of all views;and performing clustering algorithms. The two-step process cannot obtain optimal results since the two steps cannot negotiate with each other. To address this drawback, a novel algorithm named as multi-view fuzzy clustering based on anchor graph is presented. The proposed method can simultaneously obtain the membership matrix and minimize the disagreement rates of different views. A novel regularization based on trace norm is also presented in this article, which can not only obtain a clear clustering partition to prevent that all samples belonging to each cluster with the same membership value (1)/(c), but also balance the size of each cluster. Moreover, we exploit the reweighted method to optimize the proposed model, which can introduce an adaptive weight to each view to deal with the unreliable views. A series of experiments are conducted on different datasets, and the clustering performance verifies the effectiveness and efficiency of the proposed algorithm.

关键词： clustering algorithms Bipartite graph Optics Information technology Fuses Adaptation models Partitioning algorithms Fuzzy clustering graph multiview clustering reweighted optimization framework

来源：评论

学校读者我要写书评

暂无评论

Multiscale Neighborhood Adaptive clustering Image Segmentation for Molten Iron Flow Slag-Iron Recognition

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2024年第8期54卷 4642-4654页

作者： Xu, Chuan Jiang, Zhaohui Pan, Dong Yu, Haoyang Huang, Jiancai Zhou, Ke Gui, Weihua Cent South Univ Sch Automat Changsha 410083 Peoples R China

Accurately measuring the slag-iron ratio (SIR) of the molten iron flow is of great significance for efficient slag-iron emission and safe production in the ironmaking process. Slag-iron recognition is challenging due to the high temperature, intense radiation, and heavy dust in the casting field. Efficient and intelligent online slag-iron recognition is urgently needed in ironmaking enterprises. To this end, this article innovatively presents a vision-driven slag-iron recognition method based on multiscale neighborhood adaptive clustering (MNAC) to achieve continuous high-precision SIR measurement of molten iron flow. First, a shortwave infrared high-speed imaging (SWIHI) system is designed and deployed to capture high-definition molten iron flow images. Then, a multialgorithm fusion key frame and region of interest (ROI) acquisition strategy is proposed to achieve stable acquisition of molten iron flow boundaries and prevent the ROI from being occluded by dust and objects. Finally, a novel multiscale neighborhood slag-iron recognition strategy and an adaptive density peak clustering algorithm are presented to achieve high-precision slag-iron recognition through the kernel density estimation of the membership of a single pixel in multiple neighborhoods. Industrial experiments and applications indicate that the proposed method can achieve slag-iron recognition continuously and accurately to provide reliable SIR data for blast furnace experts.

关键词： Iron Slag Image segmentation Image recognition Monitoring Temperature measurement clustering algorithms ironmaking multiscale neighborhood adaptive clustering (MNAC) shortwave infrared imaging slag-iron recognition

来源：评论

学校读者我要写书评

暂无评论

Adaptive Semi-Supervised Fuzzy C-Means Method With Local Spatial Information and Pre-clustering for Image Segmentation

引用

IEEE ACCESS 2024年 12卷 196328-196346页

作者： Chen, Hao-Ran Wang, Xiao-Peng Wu, Jia-Xin Wang, Hai-Zhou Lanzhou Jiaotong Univ Sch Elect & Informat Engn Lanzhou 730070 Peoples R China

The semi-supervised fuzzy C-means clustering algorithm is an improved version of the fuzzy C-means algorithm, designed to utilize a small amount of supervised information to enhance the clustering results. However, many semi-supervised fuzzy C-means algorithms suffer from the inadequate use of supervised information and sensitivity to noise. Therefore, this study employs pre-clustering and label propagation to enhance efficiency of supervision and introduces spatial information to improve the robustness of algorithm to noise. First, preliminary clustering of the supervised information is conducted to distinguish feature differences within each cluster, allowing the supervised information to guide clustering more rationally. Second, supervised information is disseminated to pixels with similar features, enabling a small amount of supervised information to guide the clustering process effectively. Then, an objective function with adaptive weights is designed to calculate the weights of the local spatial information and supervision weights based on the local spatial information and label spatial information respectively, enhancing the flexibility of algorithm. Finally, experimental results on synthetic images and multiple real image datasets demonstrate that the proposed algorithm can accomplish most segmentation tasks and, in most cases, outperforms other algorithms.

关键词： clustering algorithms Image segmentation Noise Accuracy Image color analysis Robustness Noise measurement Linear programming Heuristic algorithms Lighting Fuzzy c-means semi-supervised clustering image segmentation label propagation local spatial information

来源：评论

学校读者我要写书评

暂无评论

Compound Weakly Supervised clustering

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2024年 33卷 957-971页

作者： Tao, Hong Jiang, Jiacheng Hou, Chenping Luo, Tingjin Fan, Ruidong Zhang, Jing Natl Univ Def Technol NUDT Coll Sci Changsha 410073 Hunan Peoples R China

clustering is a fundamental and important step in many image processing tasks, such as face recognition and image segmentation. The performance of clustering can be largely enhanced if relevant weak supervision information is appropriately exploited. To achieve this goal, in this paper, we propose the Compound Weakly Supervised clustering (CSWC) method. Concretely, CSWC incorporates two types of widely available and easily accessed weak supervision information from the label and feature aspects, respectively. To be specific, at the label level, the pairwise constraints are utilized as a kind of typical weak label supervision information. At the feature level, the partial instances collected from multiple perspectives have internal consistency and they are regarded as weak structure supervision information. To achieve a more confident clustering partition, we learn a unified graph with its similarity matrix to incorporate the above two types of weak supervision. On one hand, this similarity matrix is constructed by self-expression across the partial instances collected from multiple perspectives. On the other hand, the pairwise constraints, i.e., must-links and cannot-links, are considered by formulating a regularizer on the similarity matrix. Finally, the clustering results can be directly obtained according to the learned graph, without performing additional clustering techniques. Besides evaluating CSWC on 7 benchmark datasets, we also apply it to the application of face clustering in video data since it has vast application potentiality. Experimental results demonstrate the effectiveness of our algorithm in both incorporating compound weak supervision and identifying faces in real applications.

关键词： Face recognition Compounds Image reconstruction clustering methods clustering algorithms Task analysis Laplace equations Compound weak supervision constrained clustering incomplete view learning

来源：评论

学校读者我要写书评

暂无评论

Centroid-Free K-Means With Balanced clustering

引用

IEEE SIGNAL PROCESSING LETTERS 2025年 32卷 1191-1195页

作者： Meng, Bin Li, Fangfang Yang, Fan Gao, Quanxue Jilin Police Coll Dept Informat Secur Technol Changchun 130117 Peoples R China Xidian Univ Sch Telecommun Engn Xian 710071 Peoples R China Changchun Humanities & Sci Coll Dept Chinese Language & Literature Changchun 130117 Peoples R China

Currently, a wide array of clustering algorithms have emerged, yet many approaches rely on K-means to detect clusters. However, K-means is highly sensitive to the selection of the initial cluster centers, which poses a significant obstacle to achieving optimal clustering results. Moreover, its capability to handle nonlinearly separable data is less than satisfactory. To overcome the limitations of traditional K-means, we draw inspiration from manifold learning to reformulate the K-means algorithm into a new clustering method based on manifold structures. This method not only eliminates the need to calculate centroids in traditional approaches, but also preserves the consistency between manifold structures and clustering labels. Furthermore, we introduce the l(2,1)-norm to naturally maintain class balance during the clustering process. Additionally, we develop a versatile K-means variant framework that can accommodate various types of distance functions, thereby facilitating the efficient processing of nonlinearly separable data. The experimental results of several databases confirm the superiority of our proposed model.

关键词： clustering algorithms Manifolds Manifold learning Signal processing algorithms Optimization Linear matrix inequalities Kernel Training Geometry Data mining clustering l(2,1)-norm manifold learning K-means

来源：评论

学校读者我要写书评

暂无评论

Exploring Evolutionary Spectral clustering for Temporal-Smoothed Clustered Cell-Free Networking

引用

IEEE WIRELESS COMMUNICATIONS LETTERS 2025年第2期14卷 494-498页

作者： Wang, Junyuan Wu, Tianyao Zhou, Ouyang Zhu, Yaping Tongji Univ Coll Elect & Informat Engn Shanghai 201804 Peoples R China Tongji Univ Sch Comp Sci & Technol Shanghai 201804 Peoples R China

Clustered cell-free networking, which dynamically partitions the whole network into nonoverlapping subnetworks, has been recently proposed to mitigate the cell-edge problem in cellular networks. However, prior works only focused on optimizing clustered cell-free networking in static scenarios with fixed users. This could lead to a large number of handovers in the practical dynamic environment with moving users, seriously hindering the implementation of clustered cell-free networking in practice. This letter considers user mobility and aims to simultaneously maximize the sum rate and minimize the number of handovers. By transforming the multi-objective optimization problem into a time-varying graph partitioning problem and exploring evolutionary spectral clustering, a temporal-smoothed clustered cell-free networking algorithm is proposed, which is shown to be effective in smoothing network partitions over time and reducing handovers while maintaining similar sum rate.

关键词： Handover Optimization clustering algorithms Cellular networks Partitioning algorithms Transforms Uplink Signal to noise ratio Precoding Heuristic algorithms Clustered cell-free networking user mobility temporal smoothness handover evolutionary spectral clustering

来源：评论

学校读者我要写书评

暂无评论

A Novel clustering Routing Algorithm for Bridge Wireless Sensor Networks Based on Spatial Model and Multicriteria Decision Making

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第16期11卷 27775-27789页

作者： Yang, Jiguang Huo, Jiuyuan Mu, Cong Lanzhou Jiaotong Univ Sch Elect & Informat Engn Lanzhou 730070 Peoples R China Chinese Acad Sci Northwest Inst Ecoenvironm & Resources Natl Cryosphere Desert Data Ctr Lanzhou 730000 Peoples R China

The bridge health monitoring system (BHMS) has been widely implemented and advocated globally to replace traditional bridge management and maintenance practices. Employing wireless sensors for collecting and transmitting bridge monitoring data via a self-organizing bridge wireless sensor network (BWSN) is crucial for ensuring the efficient operation and cost-effectiveness of BHMS. To validate and enhance the performance of BWSN, this study develops a bridge space model (SM-SSB) based on the Simple-supported Beam Bridge, and introduces a critical-TOPSIS clustering (CTC) routing algorithm on SM-SSB, employing multicriteria decision making (MCDM) for optimal cluster head (Optimal-CH) and relay node (Optimal-RN) selection efficiently. The CTC algorithm's main benefit is its holistic approach to considering factors, such as the distance between nodes and the base station (BS), node density, node residual energy, and the frequency of CH selection. Utilizing the technique for order preference by similarity to an ideal solution (TOPSIS) algorithm enhanced by the criteria importance though intercrieria correlation (CRITIC) method (referred to as CRITIC-TOPSIS) sorting all nodes and then proceeds to quickly opt for the Optimal-CH and the Optimal-RN within the subspace of each bridge space model simultaneously. In the MATLAB environment, the CTC algorithm's performance is benchmarked against LEACH, Fuzzy-TOPSIS, CTC, EEM-CRP, and adaptive remora optimization algorithm, examining metrics like dead node distribution, network life cycle, packet reception rate, and variance of energy consumption. Simulation outcomes indicate that the CTC algorithm achieves a balanced energy consumption across nodes, significantly boosting the energy efficiency of bridge sensor networks and fulfilling the BWSN routing algorithm's conditions.

关键词： Bridges clustering algorithms Wireless sensor networks Routing Monitoring Sensors Optimization Bridge space model bridge wireless sensor network (BWSN) cluster routing algorithms criteria importance though intercrieria correlation (CRITIC) optimal cluster head (Optimal-CH) technique for order preference by similarity to an ideal solution (TOPSIS)

来源：评论

学校读者我要写书评

暂无评论

A Probabilistic Approach for Extractive Summarization Based on clustering Cum Graph Ranking Method

引用

IEEE ACCESS 2024年 12卷 70464-70479页

作者： Ahmad, Amreen Ahmad, Tanvir Masood, Sarfaraz Siddiqui, Mohd Khizir Abd El-Rahiem, Basma Plawiak, Pawel Alblehai, Fahad Galgotia Coll Engn & Technol Greater Noida 201310 India Jamia Millia Islamia Dept Comp Engn New Delhi 110025 India BITS Pilani Dept Comp Engn Pilani 333031 Rajasthan India Menoufia Univ Fac Sci Dept Math & Comp Sci Shibin Al Kawm 32511 Egypt Cracow Univ Technol Fac Comp Sci & Telecommun Dept Comp Sci PL-31155 Krakow Poland Polish Acad Sci Inst Theoreticl & Appl Informat PL-44100 Gliwice Poland King Saud Univ Community Coll Comp Sci Dept Riyadh 11437 Saudi Arabia

Online information has increased tremendously in today's age of the Internet. As a result, the need has arisen to extract relevant content from the plethora of available information. Researchers are widely using automatic text summarization techniques for extracting useful and relevant information from voluminous available information. The summary obtained from the automatic text summarization often faces the issues of diversity and information coverage. Earlier researchers have used graph-based approaches for ranking and optimization. This research work introduces a probabilistic approach named as ClusRank for summary extraction, comprising of a two-stage sentence selection model involving clustering and then ranking of sentences. The initial stage involves clustering of sentences using a proposed overlapping clustering algorithm on the weighted network, and later selection of salient sentences using the introduced probabilistic approach. In the analysis of real-world networks, community structure development is essential because it provides strategic insights that help decision-makers make well-informed choices. Furthermore, methodologically strict community detection algorithms are required due to the occurrence of discontinuous, overlapping, and nested community patterns in such *** research work, an algorithm is presented for detecting overlapping communities based on the concept of rough set and granular information on links. The sentence selection algorithm based on budget maximum coverage approach supports the assumption that larger sub-topics in a document are of more importance than smaller subtopics. The performance of the proposed probabilistic ClusRank is validated on DUC2001, DUC 2002, DUC2004, and DUC 2006 data sets.

关键词： Data mining clustering algorithms Electronic mail Probabilistic logic Feature extraction Rough sets Optimization Text processing Ranking (statistics) Information processing Automatic text summarization clustering graph ranking diversity information coverage

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：