检索结果-内蒙古大学图书馆

IEEE International Conference on Image Processing

作者： Sen Jia Yuntao Qian Jiming Li Yan Li Zhong Ming Shenzhen City Key Laboratory of Embedded System Design College of Computer Science and Software Engineering Shenzhen University Shenzhen China College of Computer Science University of Zhejiang Hangzhou China

In this paper, we not only extend the temporal hierarchical alternating least squares (HALS) to spatial domain, but also incorporate two necessary characteristics of material abundances, full additivity and sparsity, to unmix hyperspectral data. The new algorithm is abbreviated as HALSSC (HALS with Sparsity Constraint). Different from the other endmember extraction approaches, the proposed algorithm does not need the existence assumption of pure pixel of each endmember in the scene. Experimental results on highly mixed synthetic data and real hyperspectral data from Washington DC mall confirm the accuracy of the developed algorithm.

关键词： Hyperspectral imaging Pixel Signal processing algorithms Data mining Reflectivity Materials

来源：评论

学校读者我要写书评

暂无评论

Feature Selection Technique for Hyperspectral Imagery Classification with Noise Reduction Preprocessing

Feature Selection Technique for Hyperspectral Imagery Classi...

引用

Chinese Conference on Pattern Recognition (CCPR)

作者： Sen Jia Zhen Ji Zexuan Zhu Yuntao Qian Shenzhen City Key Laboratory of Embedded System Design College of Computer Science and Software Engineering Shenzhen University Shenzhen China College of Computer Science University of Zhejiang Hangzhou China

The rich information available in hyperspectral imagery has posed significant opportunities for material classification and identification. The main problem encountered with the classification process is the high dimensionality of hyperspectral data and the low-sized training dataset. Hence, dimensionality reduction is often adopted to avoid the "curse of dimensionality" phenomenon. However, noise generated by various sources (primarily the sensor and the atmosphere) inevitably decrease the precision of the classifier. In this paper, two wavelet-based methods, wavelet shrinkage and discrete wavelet transform, are applied to preprocess the hyperspectral imagery in sequence, denoising the spatial images and spectral signatures, respectively. After that, affinity propagation, which is a recently proposed feature selection approach, is used to choose representative features from the noise-reduced data. Experimental results demonstrate that the features acquired by the new scheme make the classification results more accurate than those without noise reduction preprocessing.

关键词： Hyperspectral imaging Discrete wavelet transforms Noise reduction Noise Accuracy

来源：评论

学校读者我要写书评

暂无评论

Feature extraction and selection hybrid algorithm for hyperspectral imagery classification

Feature extraction and selection hybrid algorithm for hypers...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Sen Jia Yuntao Qian Jiming Li Weixiang Liu Zhen Ji Shenzhen City Key Laboratory of Embedded System Design College of Computer Science and Software Engineering Shenzhen University Shenzhen China College of Computer Science University of Zhejiang Hangzhou China

Due to the enormous amounts of data contained in hyperspectral imagery, the main challenge for hyperspectral image classification is to improve the accuracy with less computation complexity. Hence, dimensionality reduction (DR) is often adopted, which includes two different kinds of methods, feature extraction and feature selection. In this paper, discrete wavelet transform (DWT) and affinity propagation (AP), which belong to feature extraction and feature selection respectively, are combined together to accomplish the DR task. Firstly, DWT-based features are extracted from the original hyperspectral data; secondly, AP is applied to select representative features from the obtained ones. Experimental results demonstrate that, compared with some other DR methods which only make use of feature extraction or feature selection, the features acquired by the hybrid technique make the classification results more accurate.

关键词： Feature extraction Hyperspectral imaging Discrete wavelet transforms Accuracy Approximation methods Noise

来源：评论

学校读者我要写书评

暂无评论

引用

International Conference on Educational and Information Technology (ICEIT)

作者： Senshen Hao Jian Jiao Chunhe Xia Xiaojian Li Shan Yao State Key Laboratory of Virtual Reality Technology and System Beijing Key Laboratory of Network Technology School of Computer Science and Engineering Beihang University Beijing China

ISBN: (纸本)9781424480333

In order to effectively deal with large-scale attacks on computer networks, computer Network Defense (CND) policy refinement based on descriptive language is wildly used. However, it's very difficult to figure out the semantic discrepancies between the measures and the predefined policy after the calculation with symbols by computers. A new method is presented to solve the discordance of semantic between the measures and the predefined policy automatically. Based on the establishment of the ontology of CND policy and measure (CNDPM), the CND policy and measure semantic similarity analysis model (CNDPMSSAM) is established, and then the termination of the two main components of CNDPMSSAM are proofed by the putdown automaton, and the prototype system of CNDPMSSAM is implemented. At last, we validate the validity of this method on analyzing semantic similarity of transferring from computer Network Defense Policy Specification Language (CNDPSL) to Defense Measure Description Language (DMDL) with experiments.

关键词： CND measure CND policy ontology semantic similarity Ontology Semantics Policies ANALYSIS MODELS Discordance Signs and symbols

来源：评论

学校读者我要写书评

暂无评论

Descriptive model of peer-to-peer Botnet structures

Descriptive model of peer-to-peer Botnet structures

引用

International Conference on Educational and Information Technology (ICEIT)

作者： Junfeng Duan Jian Jiao Chunhe Xia Shan Yao Xiaojian Li State Key Laboratory of Virtual Reality Technology and System Beijing Key Laboratory of Network Technology School of Computer Science and Engineering Beihang University Beijing China

ISBN: (纸本)9781424480333

Peer-to-peer (P2P) Botnets, which are more resilient and robust than centralized botnets, have emerged as the peer-to-peer technology evolves. Better understanding of this new phenomenon will help researchers develop detection and mitigation methods. Most of existing work is case study of typical P2P botnets. In this paper, we focus on systematically analyzing structures of P2P botnets. We propose a descriptive model of P2P botnet structures, which consists of features of P2P bot, definitions of structures, and structural properties. Firstly, we detail two key functionalities of P2P bot, command-and-control (C&C) functionality and peer-to-peer (P2P) functionality, and give several features of P2P bot. And then, we define two structures of P2P botnets: C&C structure and P2P structure. To characterize these structures, we propose several properties and corresponding quantitative methods. Finally, we conduct experiments to verify our results.

关键词： Peer to peer computing Routing protocols Concrete Storms Analytical models Fires

来源：评论

学校读者我要写书评

暂无评论

Fan-in sensitive low power dynamic circuits performance statistical characterization

Fan-in sensitive low power dynamic circuits performance stat...

引用

IEEE International SOC Conference

作者： Jinhui Wang Na Gong Wuchen Wu Ligang Hou VLSI and System Laboratory Berlin University of Technology Beijing China Department of Computer Science and Engineering SUNY at Buffalo Buffalo NY USA

Using neural networks, a highly reliable and precise system to display low power dynamic circuit performance statistical characterization is proposed in this paper. The proposed model successfully estimates the nonlinear changing of the leakage power, the active power and the delay of the different fan-in low power dynamic gates with the dual threshold voltage technique, the multiple-supply technique and the sleep transistor technique. At last, the precision priority of estimating system is obtained.

关键词： Delay Artificial neural networks Transistors Very large scale integration Estimation Logic gates Clocks

来源：评论

学校读者我要写书评

暂无评论

A Fine-Grained Data Reconstruction Algorithm for Solid-State Disks

A Fine-Grained Data Reconstruction Algorithm for Solid-State...

引用

International Conference on Networking, Architecture, and Storage (NAS)

作者： Peng Wang Diqing Hu Changsheng Xie Jianzong Wang Xiao Qin Wuhan National Laboratory of Optoelectronics Key Laboratory of Data Storage System Ministry of Education School of Computer Science and Technology Huazhong University of Science and Technology Wuhan Hubei China Department of Computer Science and Software Engineering Samuel Ginn College of Engineering Aubum University USA

Solid-state disks (SSDs) with high I/O performance are increasingly becoming popular. To extend the life time of flash memory, one can apply wear-leveling strategies to manage data blocks. However, wear-leveling strategies certainly inevitably degrade write performance. In addition to low write performance, wear-leveling strategies make one block unwritable when one bit of this block is invalid. Although data reconstruction techniques have been widely employed in disk arrays, the reconstruction techniques has not been studied in the context of solid-state disks. In this paper, we present a new fine-grained data-reconstruction algorithm for solid-state disks. The algorithm aims to provide a simple yet efficient wear-leveling strategy that improves both I/O performance and reliability of solid-state disks. Simulation experiments show that all data blocks have very similar in terms of erasure times. The number of extra erasures incurred by our algorithm is very marginal.

关键词： Reconstruction algorithms Reliability Mathematical model Flash memory Equations Arrays Parity check codes

来源：评论

学校读者我要写书评

暂无评论

Feature selection for semi-supervised multi-label learning with application to gene function analysis 10

Feature selection for semi-supervised multi-label learning w...

引用

Proceedings of the First ACM International Conference on Bioinformatics and Computational Biology

作者： Li, Guo-Zheng You, Mingyu Ge, Lei Yang, Jack Y. Yang, Mary Qu MOE Key Laboratory of Embedded System and Service Computing Tongji University Shanghai 201804 China Department of Control Science and Engineering Tongji University Shanghai 201804 China School of Computer Science and Engineering Shanghai University Shanghai 200072 China Purdue Univ. ECE West Lafayette IN 47907 United States Indiana Univ. Bioinformatics Center Indianapolis IN 46202 United States

ISBN: (纸本)9781450304382

This paper investigates gene function annotation of Yeast by using semi-supervised multi-label learning. Multi-label learning has been a hot topic in the bioinformatics field, but there are many samples unlabeled. Semi-supervised learning may be employed to utilize the unlabeled data. This paper proposes a novel semi-supervised multi-label learning algorithm COMN by combining Co-Training with ML-kNN to utilize the unlabeled yeast gene data to improve modeling accuracy of function annotation. Furthermore, an embedded feature selection algorithm PRECOMN is proposed to perform feature selection for COMN to remove the irrelevant and redundant features. Experimental results on one benchmark data set of Yeast show COMN and PRECOMN perform better than the original multi-label learning algorithm ML-kNN. Furthermore PRECOMN improves generalization performance of COMN. Copyright © 2010 ACM.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Secure Mobile Services by Face and Speech Based Personal Authentication

Secure Mobile Services by Face and Speech Based Personal Aut...

引用

2010 IEEE International Conference on Intelligent Computing and Intelligent systems(2010 IEEE 智能计算与智能系统国际会议 ICIS 2010)

作者： Linlin Shen Songhao Zheng Wei Li Nengheng Zheng Shenzhen Key Laboratory of Embedded System Design College of Computer Science & Software Engineerin College of Information & Engineering Shenzhen University Shenzhen China

With the popularity of mobile devices, people are now able to access mobile services like website surfing, e-banking, e-shopping and voice-mail etc. anywhere anytime. Most of the services require secure biometric based personal authentication. This paper aims to improve the accuracy of personal authentication in mobile environment by integrating multiple modal biometrics, i.e. face and speech. A specially designed Mobile Biometry (MOBIO) database was used to test the performances. While facial biometrics achieved an average EER of 27.3％, speech biometrics achieved an average EER of 33.4％. The multiple modals biometric system reduces the average error rate to 21.0％, which clearly proves the usefulness of fusing face and speech information.

关键词： Face recognition speech recognition fusion

来源：评论

学校读者我要写书评

暂无评论

The role of Ga(Gallium)-flux and AlN(Aluminum Nitride) as the interface materials, between (Ga-face)GaN and (Siface) 4H-SiC, through Molecular Dynamics Simulation

引用

World Academy of science, engineering and Technology 2010年 46卷 488-491页

作者： Bose, Srikanta Mazumder, Sudip K. Laboratory for Energy and Switching-Electronics System Department of Electrical and Computer Engineering University of Illinois at Chicago Science and Engineering Office 851 South Morgan Street Chicago IL: 60607-7053 United States

We report here, the results of molecular dynamics simulation of p-doped (Ga-face)GaN over n-doped (Siface)(0001)4H-SiC hetero-epitaxial material system with one-layer each of Ga-flux and (Al-face)AlN, as the interface materials, in the form of, the total Density of States (DOS). It is found that the total DOS at the Fermi-level for the heavily p-doped (Ga-face)GaN and ndoped (Si-face)4H-SiC hetero-epitaxial system, with one layer of (Al-face)AlN as the interface material, is comparatively higher than that of the various cases studied, indicating that there could be good vertical conduction across the (Ga-face)GaN over (Si-face)(0001)4HSiC hetero-epitaxial material system.

关键词： Silicon carbide

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：