检索结果-内蒙古大学图书馆

12th International Conference on CYBER technology in Automation, Control, and Intelligent systems, CYBER 2022

作者： Luo, Zhehao Luo, Sijin Liang, Guoyuan Wu, Xinyu Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China University of Chinese Academy of Sciences Beijing China Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Chinese Academy of Sciences Guangdong Province Shenzhen China Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

Depth estimation is an essential task for understanding the geometry of 3D scenes. Compared with multi-view-based methods, monocular depth estimation is more challenging for the requirement of integrating not only global information but also local cues from various parts. Recently, numerous Convolutional Neural Networks (CNNs) based approaches relying on pre-defined static architectures have been reported. Although many of them have achieved remarkable improvements in accuracy, there remains problem with the generality of the static model when it comes to diverse scenes. In this paper, a novel deep network with dynamic routing architecture is proposed with a dynamically adjustable structure with respect to different inputs, therefore leading to not only improvements on both accuracy and generality but also decrease of the model parameters. Extensive experiments of monocular depth estimation have been conducted on two well-known challenging datasets (i.e., NYUD-v2 and Make3D) as well as the mixed dataset, and have verified the effectiveness of the proposed dynamic framework. © 2022 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Occlusion avoidance approach based on depth image of visual object

引用

ICIC Express Letters, Part B: Applications 2015年第11期6卷 3103-3108页

作者： Zhang, Shihui Sang, Yu Zhang, Xiaojun School of Information Science and Engineering Yanshan University No. 438 Hebei Avenue Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 Hebei Avenue Qinhuangdao China E and A College Hebei Normal University of Science and Technology No. 360 West Hebei Street Qinhuangdao China

Occlusion avoidance is a more challenging problem in visual field. An occlusion avoidance approach based on depth image of visual object is proposed in this paper. Firstly, the occlusion detection is implemented for the depth image of visual object in current view, and then the occlusion boundary and nether adjacent boundary are obtained to construct occlusion region. Secondly, the occlusion region is segmented into several sub-regions by connecting the corner of occlusion boundary with its nether adjacent boundary point, and the triangular mesh model of each sub-region is established. Thirdly, the candidate observation direction set is determined by utilizing the normal vector of each triangle in triangular mesh model. Finally, the next best view is confirmed based on the candidate observation direction set to realize the occlusion avoidance function. Compared with the existing methods, the proposed approach does not need the priori knowledge of visual object or limit the camera position on a fixed surface. Experimental results demonstrate its feasibility and effectiveness. © 2015 ICIC International.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

A study on the influence propagation model in topic attention networks

引用

International Journal of Performability Engineering 2017年第5期13卷 721-730页

作者： Chen, Xiao Guo, Jingfeng Tian, Kelun Fan, Chaozhi Pan, Xiao College of Information Science and Engineering YanShan University Qinhuangdao066004 China Qian'An College North China University of Science and Technology Qian'an064400 China College of Economic and Management Shijiazhuang Tiedao University Shijiazhuang050043 China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao066004 China

The social networks with the complex user relations and huge amount of data and hidden information, bring new opportunities and challenges for the study of information diffusion and influence maximization. In recent years, there are more and more researches on the influence maximization of topic preference. However, most of the existing researches only take the topic as an attribute of the users, and the importance of the topic in network structure is not considered. In view of this situation, firstly, this paper constructed a new topic attention network model fusing the social relation and the topic preference. Secondly, based on connected degree of set pair and Markov random walk model, we propose the calculated method of the topic preference for users, and then mining the seed set with influence by the greedy strategy. Thirdly, we propose the calculated method of the activation probability of the user based on the user relation and the topic preference, and propose the influence maximization algorithm TAN-CELF in topic attention networks. Finally, on Dou-ban network dataset, from three metrics ISST, ISRT and ISRNT, compare with algorithm L-GAUP and CELF, the experimental results show that algorithm TAN-CELF that is proposed by this paper has a higher performance on influence scope. © 2017 Totem Publisher, Inc. All rights reserved.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Skin panel design for a morphing mechanism

引用

ICIC Express Letters, Part B: Applications 2016年第10期7卷 2161-2167页

作者： Zhao, Qiang Guo, Xliuan Xi, Fengfeng Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province College of Information Science and Engineering Yanshan University No. 438 Hebei Avenue Qinhuangdao066004 China Department of Aerospace Engineering Ryerson University TorontoONM5B 2K3 Canada

A segmentation method is proposed to aid rigid skin panel design for an existing modular morphing wing mechanism. The method divides the surface of the modular morphing mechanism into panels through geometry discretization, and then the skin which is made up of these rigid panels will have better rigidity and it will morph more free without any gap. The method comprises three basic morphing types based on the motion of upper platform. Wherein, the mechanism side is firstly divided into a plurality of panels and each of them is approximated to flat, and then shape changes of one panel before and after morphing are analyzed. Therefore, panels can be designed as moveable and fixed patches to achieve seamless morphing. At last, computer simulation is set up to further verify the correctness of the method. © 2016 ICIC International.

关键词： Morphing mechanism Seamless morphing Skin panel Surface segmentation

来源：评论

学校读者我要写书评

暂无评论

Frequent closed itemsets mining using ITBitree

引用

International Journal of Advancements in Computing technology 2012年第17期4卷 271-279页

作者： Ren, Jiadong Song, Wei Yu, Shiying College of Information Science and Engineering Yanshan University Qinhuangdao Hebei 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao City 066004 China Hebei Provincial S and T Management Information Center 159 Dongfeng Road Shijiazhuang Hebei 050021 China

The set of frequent closed itemsets is much smaller than the one of all frequent itemsets. Meanwhile it retains the completeness. In this paper, we propose an ITBitree-based algorithm, called ITBitreeFCIM (ITBitree Frequent Closed Itemsets Miner), for mining the frequent closed itemsets directly. A novel structure termed ITBitree (Itemset Tidset Binary Tree) is defined to store the transaction and itemset information. Frequent closed itemsets are generated directly by searching the right nodes in each layer and the right nodes in their respective search space deeply from this tree using top-down strategy. For the items with same support in the adjacent layers or the items with the same Tids, we use item merging technique to prune the search space. The searched space is used to avoid the generation of duplicates and enormous closure checks. In the process we do not need to check whether a candidate itemset is closed, so we save a lot of time. The experiment proofs the efficiency of the proposed approach.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Dynamic spectrum allocation scheme with real-time traffic and its performance optimization

引用

ICIC Express Letters, Part B: Applications 2015年第8期6卷 2243-2250页

作者： Shi, Guoying Jin, Shunfu Li, Yongqiang School of Information Science and Engineering Yanshan University No. 438 Hebei Avenue Qinhuangdao China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 Hebei Avenue Qinhuangdao China The First Hospital of Qinhuangdao No. 258 Wenhua Road Qinhuangdao China

Aiming at relieving the shortage problem of wireless frequency resource, sec- ondary users in cognitive radio networks occupy the spectrum hole opportunistically. In cognitive radio networks, primary users have preemptive priority for the licensed spectrum, and the data transmission of a secondary user is possible to be interrupted by a primary user at any time. For the applications with real-time traffic, the interrupted data is supposed to be discarded by the system. A continuous time queueing model with potential transmission interruption is established accordingly. By using the Laplace transform, the performance measures, such as the average latency, the normalized throughput and the spectrum switching ratio, are presented. The individually and socially optimal strategies are investigated with a non-cooperative game theory, and then a pricing policy for the secondary users is proposed. Experimental results show that the dynamic spectrum allocation scheme with real-time traffic can be optimized socially. © 2015, ICIC International.

关键词： Laplace transforms

来源：评论

学校读者我要写书评

暂无评论

Mining critical nodes in software execution network based on complex network

引用

ICIC Express Letters, Part B: Applications 2016年第9期7卷 1879-1884页

作者： Wang, Lei Dong, Jun Ren, Jiadong College of Information Science and Engineering Yanshan University No. 360 Hebei Ave Qinhuangdao066004 China E & A College Hebei Normal University of Science and Technology No. 360 Hebei Ave Qinhuangdao066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 Hebei Ave Qinhuangdao066004 China

It is significant for measuring the importance of nodes accurately to improve software stability and robustness in software network. A software execution directed network takes function as a node and relationship of function as an edge in this paper. To nd the critical nodes in the network, a novel method is proposed to measure the nodes' importance by means of depth search mining in software execution. According to the principle of cascading failure, a novel critical nodes metrics FID is dened for nodes measuring and sorting. Critical nodes mining (CNM) algorithm is put forward for calculating FID value of each node and sorting. We choose top-ranking nodes as critical nodes which play an important role in software execution process. Experimental results show that CNM algorithm can measure the critical nodes accurately in software network. © 2016 ICIC International.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Mobile APP user attribute prediction by heterogeneous information network modeling

arXiv

引用

arXiv 2019年

作者： Zhang, Hekai Gong, Jibing Teng, Zhiyong Wang, Dan Wang, Hongfei Du, Linfeng Bhuiyan, Zakirul Alam School of Information Science and Engineering Yanshan University Qinhuangdao066004 China Key Lab for Computer Virtual Technology and System Integration Yanshan University Qinhuangdao066004 China School of Computer Science and Engineering Beihang University Beijing100191 China Shenyuan Honors College Beihang University Beijing100191 China State Key Lab of Mathematical Engineering and Advanced Computing Wuxi214000 China Department of Computer and Information Sciences Fordham University

User-based attribute information, such as age and gender, is usually considered as user privacy information. It is difficult for enterprises to obtain user-based privacy attribute information. However, user-based privacy attribute information has a wide range of applications in personalized services, user behavior analysis and other aspects. Although many scholars have made achievements in user attribute prediction and other related fields, there are still two main problems that impede further improvement on the accuracy of classification: (1) Traditional machine learning classification merely takes each object as a single individual, ignoring the relationship between them;(2) At present, the popular Heterogeneous Path-Mine Information Network only considers whether the user has a relationship with the attributes of other nodes, rather than the degree of correlation of the attributes. It employs a linear regression model to fit the weight of meta-path, which is highly sensitive to outliers. To solve the above two problems, this paper advances the HetPathMine model and puts forward TPathMine model. With applying the number of clicks of attributes under each node to express the user's emotional preference information, optimizations of the solution of meta-path weight are also presented. Based on meta-path in heterogeneous information networks, the new model integrates all relationships among objects into isomorphic relationships of classified objects. Matrix is used to realize the knowledge dissemination of category knowledge among isomorphic objects. The experimental results show that: (1) the prediction of user attributes based on heterogeneous information networks can achieve higher accuracy than traditional machine learning classification methods;(2) TPathMine model based on the number of clicks is more accurate in classifying users of different age groups, and the weight of each meta-path is consistent with human intuition or the real world situation. Copyright ©

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Minkowski sum modeling of polyhedra based on VRML

引用

ICIC Express Letters 2013年第2期7卷 455-460页

作者： Zhang, Jianfei Guo, Xijuan Geng, Qingjia Information Science and Engineering College Yanshan University No. 438 Hebei Ave. Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of HeBei Province No. 438 Hebei Ave. Qinhuangdao 066004 China Computer and Control Engineering College Qiqihar University No. 42 Wenhua Street Qiqihar 161006 China

On the basis of the introduction to basic methods of the geometry modeling based on virtual reality modeling language (VRML) and the concepts of Minkowski sum computation of polyhedra, we present a method of Minkowski sum computation and three dimensional modeling based on VRML. The method can simplify Minkowski sum computation of general polyhedra, and experimental results indicate that this method can calculate Minkowski sum of general polyhedra much faster and realize three-dimensional representation more accurately. This method can provide important technical references for Minkowski sum visualization of polyhedra. © 2013 ISSN 1881-803X.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

RNA secondary structure prediction based on density-based clustering

引用

ICIC Express Letters 2015年第7期9卷 1965-1970页

作者： Wang, Changwu Wang, Xiuqin Wang, Baowen Liu, Wenyuan Wei, Zhenzhen Li, Yongqiang School of Information Science and Engineering Yanshan University No. 438 Hebei Ave Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 Hebei Ave Qinhuangdao China Department of Information The First Hospital of Qinhuangdao No. 258 Hebei Ave Qinhuangdao China

The ribonucleic acid (RNA) secondary structure prediction has designed to characterize the ensemble of structures, instead of only computing the minimum free energy structure. Clustering methods are brought to aid in the characterization of the structure ensemble. These methods group the structures into a small number of clusters. A representative structure for each cluster is computed to characterize the central tendency for the structures in each cluster and used to predict the true secondary structure. This paper introduces a density based clustering with extensible radius dubbed ER-DBSCAN for handling varying density clustering. The basic idea of ER-DBSCAN is to use different values of radius for different densities. It starts from the highest density object towards the lowest density object, and the radius of the cluster is automatically adjusted during cluster expansion according to the density distribution and density variation. Our results indicate that ER-DBSCAN can detect and handle clusters of varying density, and cluster RNA secondary structures effectively. © 2015, ISSN 1881-803X.

关键词： RNA

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：