检索结果-内蒙古大学图书馆

25th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2017

作者： Geng, Yanyan Liang, Ru-Ze Li, Weizhi Wang, Jingbin Liang, Gaoyuan Xu, Chenhao Wang, Jing-Yan Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou215006 China King Abdullah University of Science and Technology Thuwal23955 Saudi Arabia Suning Commerce RandD Center USA Inc. Palo AltoCA94304 United States Information Technology Service Center Intermediate People's Court of Linyi City Linyi China Jiangsu University of Technology Jiangsu213001 China New York University Abu Dhabi Abu Dhabi United Arab Emirates

ISBN: (纸本)9782875870391

In the machine learning problems, the performance measure is used to evaluate the machine learning models. Recently, the number positive data points ranked at the top positions (Pos@Top) has been a popular performance measure in the machine learning community. In this paper, we propose to learn a convolutional neural network (CNN) model to maximize the Pos@Top performance measure. The CNN model is used to represent the multi-instance data point, and a classifier function is used to predict the lab.l from the its CNN representation. We propose to minimize the loss function of Pos@Top over a training set to learn the filters of CNN and the classifier parameter. The classifier parameter vector is solved by the Lagrange multiplier method, and the filters are updated by the gradient descent method alternately in an iterative algorithm. Experiments over benchmark data sets show that the proposed method outperforms the state-of-the-art Pos@Top maximization methods. © ESANN 2017 - Proceedings, 25th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A survey on video-based vehicle behavior analysis algorithms

引用

Journal of Multimedia 2012年第3期7卷 223-230页

作者： Wu, Jian Cui, Zhi-Ming Chen, Jian-Ming Zhang, Guang-Ming The Institute of Intelligent Information Processing and Application Soochow University Suzhou 215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou 215006 China Jiangsu Yihe Traffic Engineering Co. Ltd Suzhou 215002 China

Analysis of the Vehicle Behavior is mainly to analyze and identify the vehicles' motion pattern, and describe it by the use of natural language. It is a considerable challenge to analyze and describe the vehicles' behavior in a complex scene. This paper first hackles the development history of the intelligent transportation system and analysis of vehicles' behavior, and then conducts an indepth analysis of current situation of vehicle behavior analysis from the video processing, video analysis and video understanding, summarizes the achieved results and the key technical problems, and prospects the future development of vehicle behavior analysis. © 2012 ACADEMY PUBLISHER.

关键词： Video signal processing

来源：评论

学校读者我要写书评

暂无评论

A Privacy Policy of P3P Based on Relational Database

A Privacy Policy of P3P Based on Relational Database

引用

International Multi-Symposiums on computer and Computational Sciences (IMSCCS)

作者： Bin Jiang Da-fung Zhang Chao Yang Guangxue Yue Fei Yu Software School Hunan University Changsha Hunan China College of Computer and Communication Hunan University Changsha Hunan China Provincial Key Laboratory of Computer Information Processing Technology Jiangsu Jiangsu Province China

P3P is an industry standard method that enables users to gain powerful control over their processing of personal information on Web sites being visited. Traditional methods of database access control places great dependence on the use application of statically defined views. In this paper, we propose the procedure by which current RDBMS can be transformed into their privacy-preserving equivalents. In addition, architectural for implementing P3P based on RDBMS is dissected and a server-centric implementation that reuses database querying technology is suggested

关键词： Data privacy Relational databases Industrial control Automatic control Guidelines Legislation Chaotic communication Educational institutions lab.ratories information processing

来源：评论

学校读者我要写书评

暂无评论

An Algorithm for 3D Image Segmentation

An Algorithm for 3D Image Segmentation

引用

International Conference on Image and Graphics (ICIG)

作者： Ding Zhi Dong Yu-ning College of Communication and Information Engineering Nanjing University of Posts and Telecommunications Nanjing China State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Jiangsu Provincial Key Lab of Computer Information Processing Technology Suzhou University Suzhou China

We propose a new 3D image segmentation method based on prediction, block-matching and partial 3D constraint in this paper. The algorithm only needs to set a few key points in the first image. We use intensity information and block-matching to optimize the initial condition, and consider the 3D object's characteristics at the same time. By using the partial 3D constraints we can get the result of 2D and 3D smoothness. Experimental results validate its usefulness in 3D image segmentation.

关键词： Image segmentation Active contours Image motion analysis Shape Educational technology information processing Prediction methods Image converters Image sequences computer graphics

来源：评论

学校读者我要写书评

暂无评论

Dual-Source CBCT for Larger Longitudinal Coverage: System Design and Image Reconstruction 46

Dual-Source CBCT for Larger Longitudinal Coverage: System De...

引用

46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2024

作者： Zhang, Xusheng Lyu, Tianling Shi, Yijie Zhong, Xinyun Wu, Zhan Xi, Yan Wang, Shijie Chen, Yang Zhu, Wentao Southeast University School of Software Nanjing China Zhejiang Lab Hangzhou China First-Imaging Tech. Shanghai China Southeast University Laboratory of Image Science and Technology The School of Computer Science and Engineering Nanjing China Southeast University Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing The Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Nanjing China

ISBN: (纸本)9798350371499

The application of CBCT systems in intraoperative environments has become increasingly common, but concurrent CBCT systems are unsuitable for situations that require a large longitudinal imaging FoV, such as orthopedics. To increase longitudinal coverage, we developed a dual-source CBCT (DS-CBCT) system in which two ray sources are symmetrically placed along the central plane. After further analyzing its geometric characteristics, we propose an analytical reconstruction algorithm termed DT-FDK specialized for DS-CBCT, which combines cone-beam rebinning and two rays with smaller cone angles among the four conjugate rays to further suppress cone beam artifacts. The system design and reconstruction algorithm are tested on both simulated and real-scanned data. Results show that the DS-CBCT can expand the effective imaging volume by 37.1% compared to single-source CBCT systems. The reconstructed images by DT-FDK also show improved image quality compared to traditional reconstruction algorithms. © 2024 IEEE.

关键词： Cones

来源：评论

学校读者我要写书评

暂无评论

Generalized geometry-based optimal power control in wireless networks

Generalized geometry-based optimal power control in wireless...

引用

2012 9th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks, SECON 2012

作者： Wang, Wei Shin, Kang G. Zhang, Zhaoyang Wang, Wenbo Peng, Tao Department of Information Science and Electronic Engineering Zhejiang Provincial Key Laboratory of Information Network Technology Zhejiang University Hangzhou 310027 China Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI 48109-2121 United States Wireless Signal Processing and Networking Lab Beijing University of Posts and Telecommunications Beijing 100876 China

ISBN: (纸本)9781467319058

Geometry-based optimal power control was proposed in [14] to transform the power-control problem to a new geometrical problem on the position relationship between a line and some points. This scheme provides a novel visual perspective and lowers the complexity of optimization. We generalize this scheme to a larger class of power-control optimization problems so as to maximize the network utility with multiple average and peak power constraints in wireless networks. To facilitate the handling of the geometrical model, we define a subset of geometrical models with specified characteristics, called a regular geometrical model, and derive the type of power-control problems eligible for the regular geometrical model. For such a type of problems, two strategies are proposed for the construction of the regular geometrical model. Utilizing geometrical properties, we propose a novel geometry-based optimization scheme for the general power-control problem. Its computational complexity is significantly lower than the conventional algorithms. We also provide a further discussion on irregular geometrical model cases. Finally, we provide two examples of deploying the proposed geometry-based power-control scheme. © 2012 IEEE.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Assigning method for decision power based on linguisitc 2-tuple judgment matrices

Journal of Software

引用

Journal of Software 2011年第3期6卷 508-515页

作者： Zhang, Xixiang Yue, Guangxue Zheng, Xiajie Yu, Fei Business School of Jiaxing University Jiaxing China Mathematics and Information Engineering School of Jiaxing University Jiaxing China Department of Computer Science and Technology Huaihua University Huaihua China Jiangsu Provincial Key Laboratory for Computer Information Processing Technology Soochow University Soochow China

Decision power is very important in group decision making, which effects the final decision making result. When there exists uncertainty in group decision making, it is easy for an expert to express his/her preferences using fuzzy linguistic term such as 'good', 'very good'. The linguistic 2-tuple representation model was selected to represent fuzzy linguistic term for its accuracy in representing fuzzy terms and less loss of information in processing. To obtain the objective decision power in group decision making where the experts express their opinions using linguistic 2-tuple judgment matrices, the paper put forward a new approach to assign decision power based on element consistency level. And the method for calculating the contribution degree of each expert was also provided. The proposed method thought of calculating the different weight to aggregate different element according to the element consistent level. An illustrated example was used to demonstrate the proposed method. And a GDSS based on the method was also developed to simplify the application in ***. The GDSS was applied to building non-financial performance system for listed company in small and medium enterprise board in China, which showed its reasonability. © 2011 ACADEMY PUBLISHER.

关键词： Linguistics

来源：评论

学校读者我要写书评

暂无评论

Linear support vector machine based on kernel function

引用

Journal of Computational information Systems 2009年第4期5卷 1089-1095页

作者： Gao, Shang Hu, Xuekun Zhang, Zaiyue Cao, Cungen School of Computer Science and Engineering Jiangsu University of Science and Technology Zhenjiang 212003 China Provincial Key Laboratory of Computer Information Processing Technology Soochow University Suzhou 215006 China Institute of Computing Technology The Chinese Academy of Sciences Beijing 100080 China

Support vector machine is kind of novel machine learning method, based on statistical learning theory, which becomes the hotspot of machine learning because of its excellent learning performance. The method of support vector machine has been developed for solving classification and regression problems. Training a support vector machine requires the solutions of a very large quadratic programming problem. The algorithm is usually slow when there are many data. In order to improve the running speed, three regression models based on kernel function are introduced. One of three models is changed into a system of linear equations problem, the other models are changed into linear programming problems and their dual problems are given. Simulation results show the effectiveness of the methods, and the computation complexity are largely reduced. The kernel function and the parameter of models selection are compared and analyzed by means of an example. © 2009 Binary information Press August, 2009.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Fuzzy inference mechanism based automatic elastic registration of two-dimensional gel electrophoresis images

引用

information technology Journal 2012年第1期11卷 20-29页

作者： Pan, Yonghui Deng, Zhaohong Wang, Shitong Gao, Qun Jiangsu Engineering R and D Center for Information Fusion Software Jiangsu Jiangyin 214405 China School of Digital Media Jiangnan University Lihu Avenue 1800 Wuxi Jiangsu 214122 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Jiangsu Suzhou 215123 China

A key technique for protein analysis is the geometric alignment of two-dimensional polyacrylamide gel electrophoresis (2-D PAGE), i.e., 2-D PAGE image registration. In this study, the adaptability in elastic image registration was emphasized. According to the characteristics of 2-D gel image registration, a fuzzy-inference-rule based flexible model (FIM-FM) is proposed to model the complex transformation between 2-D gel image pairs. By introducing the concept of motion estimation, the parameter learning rules of the proposed model are derived for registration. The experiments show that the proposed algorithm is highly effective for registration of 2-D gel images and is competitive to the existing state-of-the-art algorithms. © 2012 Asian Network for Scientific information.

关键词： Image registration

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection for High-Speed Networks Based on Producing System

Intrusion Detection for High-Speed Networks Based on Produci...

引用

International Workshop on Knowledge Discovery and Data Mining (WKDD)

作者： Ken Chen Fei Yu Cheng Xu Yan Liu School of Computer &Information Engineering Hunan Agricultural University Changsha China College of Computer & Communication Hunan University Changsha China Jiangsu Provincial Key Laboratory of Computer Information Processing Technology Suzhou University Suzhou China

This paper firstly analyzed the front pattern matching algorithm of intrusion detection system, and improved on the pattern matching algorithm. Having immediately analyzed Rete algorithm and aimed at its shortage, the paper introduced the FRete net algorithm on the basis of Rete net algorithm constituting an inference machine based on FRete matching algorithm, which could infer logical suspicious facts. Then the authors discussed and designed an intrusion detection system based on high-speed networks, and finally established a simulation and experiment platform to test the performance of IDS.

关键词： Intrusion detection High-speed networks Inference algorithms computer networks Data security Pattern matching Algorithm design and analysis computer architecture Protection Data mining

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：