检索结果-内蒙古大学图书馆

Multiple object tracking based on appearance and motion graph convolutional neural networks with an explainer

Neural Computing and Applications 2024年第22期36卷 13799-13814页

作者： Zhang, Yubo Huang, Qingming Zheng, Liying School of Computer Science and Technology Harbin Engineering University 145 Nantong Street Heilongjiang Harbin150001 China School of Computer Science and Technology University of Chinese Academy of Sciences Huairou District Beijing100019 China

The tracking performance of Multi-Object Tracking (MOT) has recently been improved by using discriminative appearance and motion features. However, dense crowds and occlusions significantly reduce the reliability of these features, resulting in unsatisfied tracking performance. Thus, we design an end-to-end MOT model based on Graph Convolutional Neural Networks (GCNNs) which fuses four classes of features that characterize objects from their appearances, motions, appearance interactions, and motion interactions. Specifically, a Re-Identification (Re-ID) module is used to extract more discriminative appearance features. The appearance features from object tracklets are then averaged to simplify the proposed tracker. Then, we design two GCNNs to better distinguish objects. One is for extracting interactive appearance features, and the other is for interactive motion features. A fusion module then fuses those features, getting the global feature similarity based on which an association component calculates the MOT matching results. Finally, we semantically visualize relevant structures with the GNNExplainer for insight into the proposed tracker. The evaluation results on MOT16 and MOT17 benchmarks show that our model outperforms the state-of-the-art online tracking methods in terms of Multi-Object Tracking Accuracy and Identification F1 score which is consistent with the results from the GNNExplainer. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Developing new electrocatalysts for oxygen evolution reaction via high throughput experiments and artificial intelligence

引用

npj Computational Materials 2024年第1期10卷 1231-1238页

作者： Shaomeng Xu Zhuyang Chen Mingyang Qin Bijun Cai Weixuan Li Ronggui Zhu Chen Xu X-D.Xiang School of Materials Science and Engineering Harbin Institute of TechnologyHarbin150001China School of Materials Science and Engineering Southern University of Science and TechnologyShenzhen518055China Academy for Advanced Interdisciplinary Studies Southern University of Science and TechnologyShenzhen518055China Shenzhen Polytechnic University Shenzhen518055China

The development of non-noble metal electrocatalysts for the Oxygen Evolution Reaction(OER)is advancing towards the use of multi-element *** reveal the complex correlations of multielement OER electrocatalysts,we developed an iterative workflow combining high-throughput experiments and AI-generated content(AIGC)*** increased number of 909(compared to 145 in previous literature)universal descriptors for inorganic materials science were constructed and used as Artificial Neural Network(ANN)input.A large number of statistical ensembles with each ANN individual ensemble having a reduced number of descriptors were integrated with a new Hierarchical Neural Network(HNN)*** algorithm addresses the longstanding challenge of balancing overwhelming descriptor numbers with insufficient datasets in traditional ANN approaches to materials science *** a result,the combination of AIGC and experimental validation significantly enhanced prediction accuracy,increase the R2 values from 0.7 to 0.98 for Tafel slopes.

关键词： materials artificial evolution

来源：评论

学校读者我要写书评

暂无评论

Insights into the deformation mechanisms of an Al1Mg0.4Si alloy at cryogenic temperature:An integration of experiments and crystal plasticity modeling

引用

Journal of Materials science & technology 2024年第33期200卷 69-82页

作者： Youhong Peng Danyang Li He Wu Kesong Miao Chenglu Liu Li Wang Wei Liu Chao Xu Lin Geng Peidong Wu Guohua Fan School of Materials Science and Engineering Harbin Institute of TechnologyHarbin 150001China Laboratory for Space Environment and Physical Sciences Harbin Institute of TechnologyHarbin 150001China Key Laboratory for Light-Weight Materials Nanjing Tech UniversityNanjing 211816China

In this work,we investigated the mechanical properties and corresponding deformation mechanisms of an Al1Mg0.4Si alloy,which exhibited significantly higher strength and outstanding strain hardening capacity at 77 K compared to its counterparts at 298 *** deformation mechanisms responsible for the excellent strength-ductility synergy and extraordinary strain hardening capacity at cryogenic temperature were elucidated through a combined experimental and simulation *** results reveal the presence of numerous slip traces and microbands throughout grain surfaces during deformation at 298 K,whereas at 77 K,vague grain surfaces dominate,indicating the simultaneous operation of multiple slip *** electron microscopy(TEM)analysis using the two-beam diffraction technique demonstrates the presence of dislocations with several different Burgers vectors inside a grain at cryogenic temperature,confirming the activation of multiple slip *** accumulation of dislocations facilitated by these multiple slip systems,combined with the high dislocation density,contributes to strain hardening and remarkable uniform elongation at 77 K.A modified dislocation density-based crystal plasticity model,incorporating the effect of grain boundary hardening(GBH)and temperature,was developed to gain a better understanding of the underlying mechanisms governing alloy’s strength and *** GBH effect significantly enhances statistically stored dislocation(SSD)density and screw dislocation proportion,which promote homogeneous deformation and enhance strain hardening capacity at cryogenic *** findings deepen the understanding of plastic deformation at cryogenic temperatures and pave the way for the development of ultrahigh-performance metallic materials for cryogenic applications.

关键词： Aluminum alloy Cryogenic temperature Grain boundary hardening effect Deformation mechanism Crystal plasticity modeling

来源：评论

学校读者我要写书评

暂无评论

SNN-AAD: Active Anomaly Detection Method for Multivariate Time Series with Sparse Neural Network 28th

SNN-AAD: Active Anomaly Detection Method for Multivariate T...

引用

28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

作者： Ding, Xiaoou Liu, Yida Wang, Hongzhi Yang, Donghua Song, Yichen School of Computer Science and Technology Harbin Institute of Technology Harbin China

ISBN: (纸本)9783031306365

Anomaly detection of time series data is an important and popular problem in both research and application fields. Kinds of solutions have been developed to uncover the anomaly instances from data. However, the labelled data is always limited and costly for real applications, which adds to the difficulty of identifying various anomalies in multivariate time series. In this paper, we propose a novel active anomaly detection method with sparse neural network (SNN-AAD) to improve the accuracy and efficiency in anomaly detection for time series with limited labels. SNN-AAD is designed for two objectives: (1) to achieve sufficient generalization capacity of the model with small-size labels, and (2) to effectively reduce the human cost in active learning process. We introduce sparse neural network in training which minimizes the detection loss caused by the sparsity of labels. We improve the active anomaly detection with the design of sample selection strategy and abnormal feature order generation algorithm, which extracts the important features of instances and reduce the cost of human intelligence. Experimental results on four real-life datasets show SNN-AAD has good detection performance with limited labels, and improves the detection accuracy by 10.2%–18.2% from benchmark methods, while it achieves 12.1% human cost reduction. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Galerkin-based quasi-smooth manifold element(QSME)method for anisotropic heat conduction problems in composites with complex geometry

引用

Applied Mathematics and Mechanics(English Edition) 2024年第1期45卷 137-154页

作者： Pan WANG Xiangcheng HAN Weibin WEN Baolin WANG Jun LIANG School of Civil Engineering Central South UniversityChangsha 410083China Institute of Advanced Structure Technology Beijing Institute of TechnologyBeijing 100081China School of Science Harbin Institute of TechnologyShenzhen 518055Guangdong ProvinceChina

The accurate and efficient analysis of anisotropic heat conduction problems in complex composites is crucial for structural design and performance evaluation. Traditional numerical methods, such as the finite element method(FEM), often face a trade-off between calculation accuracy and efficiency. In this paper, we propose a quasi-smooth manifold element(QSME) method to address this challenge, and provide the accurate and efficient analysis of two-dimensional(2D) anisotropic heat conduction problems in composites with complex geometry. The QSME approach achieves high calculation precision by a high-order local approximation that ensures the first-order derivative *** results demonstrate that the QSME method is robust and stable, offering both high accuracy and efficiency in the heat conduction analysis. With the same degrees of freedom(DOFs), the QSME method can achieve at least an order of magnitude higher calculation accuracy than the traditional FEM. Additionally, under the same level of calculation error, the QSME method requires 10 times fewer DOFs than the traditional FEM. The versatility of the proposed QSME method extends beyond anisotropic heat conduction problems in complex composites. The proposed QSME method can also be applied to other problems, including fluid flows, mechanical analyses, and other multi-field coupled problems, providing accurate and efficient numerical simulations.

关键词： anisotropic heat conduction quasi-smooth manifold element(QSME) composite with complex geometry numerical simulation finite element method(FEM)

来源：评论

学校读者我要写书评

暂无评论

A Constitutive Model for Softening Behaviors of Skin Tissue

引用

Acta Mechanica Solida Sinica 2024年第5期37卷 762-770页

作者： Zhonghui Yuan Zheng Zhong School of Mathematics and Statistics Qingdao UniversityQingdao266071China School of Science Harbin Institute of TechnologyShenzhen518055China

Skin tissue is a kind of complex biological material abundant with fibers.A new constitutive model,relating macroscopic responses with microstructural fiber configuration alteration,is developed to investigate the stress softening behaviors of skin tissue observed during cyclic loading-unloading *** influential factors are introduced to describe the impact of fiber configuration change and stretch-induced *** present model achieves good agreement between predicted stress distribution of human skin and corresponding ex vivo experimental data obtained from the literature,affirming its capability to effectively capture the characteristic softening behaviors of human skin under cyclic loading conditions.

关键词： Skin tissue Constitutive model Stress softening Fibrous structures Damage

来源：评论

学校读者我要写书评

暂无评论

Theoretical and experimental studies on the electrodeposited nanocrystalline Ag coatings from a novel Ag2O-based non-cyanide electrolyte

引用

Journal of Industrial and Engineering Chemistry 2024年 131卷 432-440页

作者： Li, Ruopeng Jiang, Jie Li, Yaqiang Ren, Penghui Shen, Baicheng Qin, Jiang Zhang, Jinqiu An, Maozhong Yang, Peixia MIIT Key Laboratory of Critical Materials Technology for New Energy Conversion and Storage School of Chemistry and Chemical Engineering Harbin Institute of Technology Harbin150001 China School of Energy Science and Engineering Harbin Institute of Technology Harbin150000 China

Functional silver coatings have been widely applied in fields of electrochemical energy conversion systems, electronics industry and antibacterial materials, and exploiting new non-cyanide silver electroplating bath system is of great significance for promoting the application. In this work, a new type of cyanide-free electrodeposited silver (Ag) system was developed with Ag2O as the main salt, 5,5-dimethyl hydantoin (DMH) as the silver ion complexing agent and potassium carbonate as the conductive salt. Comprehensive quantum chemical calculations demonstrate that the new silver-based system has the solvation structure of [Ag(C5H7N2O2)2]−(5N(2)-Ag). To further, roles of additives combination (L-cysteine and 2,2-bipyridine) and their fundamental mechanism on the silver electrodeposition process was further investigated by electrochemical tests and theoretical calculations. First-Principles calculation and simulation of molecular dynamics reveal that this additive couple have coordination interactions with Ag+, while strong adsorption on the Ag (1 1 1) surface. SEM and XRD analyses showed that the combination of L-cysteine and 2,2-bipyridine achieved an ideal silver coating with good morphology and a low grain size of 22.1 nm, verifying the great potential of the novel system to prepare high-quality silver coatings. © 2023

关键词： Nanocrystals

来源：评论

学校读者我要写书评

暂无评论

Co-clustering of single-cell RNA-seq data based on weighted non-negative matrix tri-factorization combined with consensus clustering

Co-clustering of single-cell RNA-seq data based on weighted ...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Ren, Tongtong Wang, Guohua Harbin Institute of Technology School of Computer Science and Technology Harbin China

ISBN: (纸本)9798350337488

Single-cell RNA sequencing (scRNA-seq) has the ability to accurately identify cell types contained in tissues at single-cell resolution. However, the research of identifying imbalanced cell types using scRNA-seq data is still challenging. In this article, we propose scCO2, a method based on weighted non-negative matrix tri-factorization (NMTF) combined with Kmeans-based consensus clustering for the co-clustering of scRNA-seq data. Compared with several popular methods on six real scRNA-seq data with known cell types, scCO2 can achieve comparable or superior cell clustering performance to selected clustering methods. Through the case study on a real scRNA-seq dataset of the human pancreas, scCO2 could obtain good correspondence between gene clusters and cell clusters. Additionally, scCO2 also shows the ability to identify rare cell types. Moreover, by comparing the gene sets from gene clusters to existing known marker genes, we demonstrate that scCO2 has the potential to identify more underlying cell-type-specific genes, and the weights of genes learned by scCO2 could be used as the indicator of gene importance. © 2023 IEEE.

关键词： co-clustering consensus clustering imbalanced cell types non-negative matrix tri-factorization single-cell RNA sequencing

来源：评论

学校读者我要写书评

暂无评论

Multiple model PHD filter for tracking sharply maneuvering targets using recursive RANSAC based adaptive birth estimation

引用

Journal of Systems Engineering and Electronics 2024年第3期35卷 780-792页

作者： DING Changwen ZHOU Di ZOU Xinguang DU Runle LIU Jiaqi School of Astronautics Harbin Institute of TechnologyHarbin 150001China School of Electronics and Information Engineering Harbin Institute of TechnologyHarbin 150001China National Key Laboratory of Science and Technology on Test Physics and Numerical Mathematics Beijing 100076China

An algorithm to track multiple sharply maneuvering targets without prior knowledge about new target birth is proposed. These targets are capable of achieving sharp maneuvers within a short period of time, such as drones and agile *** probability hypothesis density (PHD) filter, which propagates only the first-order statistical moment of the full target posterior, has been shown to be a computationally efficient solution to multitarget tracking problems. However, the standard PHD filter operates on the single dynamic model and requires prior information about target birth distribution, which leads to many limitations in terms of practical applications. In this paper,we introduce a nonzero mean, white noise turn rate dynamic model and generalize jump Markov systems to multitarget case to accommodate sharply maneuvering dynamics. Moreover, to adaptively estimate newborn targets’information, a measurement-driven method based on the recursive random sampling consensus (RANSAC) algorithm is proposed. Simulation results demonstrate that the proposed method achieves significant improvement in tracking multiple sharply maneuvering targets with adaptive birth estimation.

关键词： multitarget tracking probability hypothesis density(PHD)filter sharply maneuvering targets multiple model adaptive birth intensity estimation

来源：评论

学校读者我要写书评

暂无评论

Cleanits-MEDetect: Multiple Errors Detection for Time Series Data in Cleanits 28th

Cleanits-MEDetect: Multiple Errors Detection for Time Serie...

引用

28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

作者： Ding, Xiaoou Song, Yichen Wang, Hongzhi Yang, Donghua Liu, Yida School of Computer Science and Technology Harbin Institute of Technology Harbin China

ISBN: (纸本)9783031306778

Data quality problems are seriously prevalent in time series data, and the data suffer from types of errors including single-point errors, continuous errors, and contextual errors. Since it is challenging to achieve high accuracy and efficiency in error detection tasks for time series data, we develop error detection system MEDetect in Cleanits, a data cleaning tool for multi-dimensional industrial time series. We propose an integration detection model for multiple errors, which holds the hierarchical variational automatic encoder as the main structure, and we propose a dimensionality reduction method for k-shape based cluster- ing algorithm, which reduces the time costs of the detection process. MEDetect is designed to allow customized error detection, and users can choose detection and repairing algorithms on their demands. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Error detection

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：