用非监督式聚类进行视频镜头分割

镜头边界检测是基于内容的视频检索首先要解决的问题。研究人员通常将镜头转换分为突变和渐变,并根据各种转换的特点采用不同的检测算法。在研究中发现,视频镜头的抽象程度与对其进行边界划分的精度相关。为此,提出采用非监督式聚类算法,按照给定的相似尺度对视频数据进行自组织和动态分析,完成层次化的镜头分割。该算法侧重于揭示视频的层次结构,能实现不同精度的视频抽象要求。     

内网访问控制策略执行风险度量

VPN网关是众多涉密内网的典型边界防护设备,其网络访问控制策略的执行情况反映了涉密内网安全状况。选取典型的基于VPN网关的涉密内网,研究探讨网络访问控制策略执行过程中的风险度量问题,将安全风险作为一个要素引入网络访问控制策略之中,对策略执行过程中主体、客体和操作等基本信息进行风险分析与度量,从而获得有利于网络访问控制策略动态优化的量化度量结果,改善涉密内网安全状况。     

基于后缀树词序列核挖掘Web文档

通过将文档表示为一棵后缀树,文章提出一种基于后缀树索引计算文档相似度的词序列核.首先根据文档的词序列构造出后缀树,然后根据后缀树词序列核计算文档间的相似度,最后利用支持向量机对文档进行分类.理论分析表明后缀树词序列核的计算只与比较文档的长度成线性关系,大大减少了序列核的计算时间.在reuters21578文档集上将后缀树词序列核与词序列核、多项式核进行比较,实验结果表明在改善速度的同时,后缀树词序列核可达到与词序列核相当的性能,优于多项式核,更适于Web文档挖掘等应用.     

基于边缘特征的异质图像目标配准方法

目标检测作为计算机视觉的重要研究课题,在视频监控和智能导航方面有重要应用.图像配准技术是刚体目标检测方法之一.异质图像间目标的纹理差异极大,为此,对于目标的可见光俯视图像与机载红外摄像机观测异质图像,通过研究基于边缘特征的配准技术,实现对地面平面目标的配准检测.笔者首先通过建立机载运动摄像机空间模型,解决了红外观测图像的视角变换问题,将其换到与参考图像一致的视角,提出一种新颖的基于稳定区域轮廓合成的边缘提取方法,更好地控制异质图像的边缘提取效果,边缘特征匹配是在距离变换的基础上实现的,提出了一种基于稳定区域距离图像互相关的匹配度量准则.实验表明,算法能有效地实现异质图像间的目标配准检测.     

软件质量模糊综合评价模型与实例分析

质量评价是软件质量管理的重要内容,针对软件质量评价指标难以量化且具有不确定性的特点,运用模糊综合评价法进行研究。首先,将软件质量评价指标体系归纳为6个一级指标,24个二级指标,并用递阶层次分析结构模型表示;接着,研究评价工作准备,包括建立因素集、建立评价集和建立权重集;然后,建立数学模型,包括单因素模糊评价、一级模糊综合评价、多级模糊综合评价和确定评价结果;最后,按照数学模型建立步骤对具体软件进行质量评价。研究内容对于加强软件项目管理、提高软件质量具有重要意义。     

数据驱动的软件缺陷预测研究综述

数据驱动的软件缺陷预测是提高软件测试效率、保证软件可靠性的重要途径之一,近几年已成为实证软件工程的研究热点.首先介绍了数据驱动软件缺陷预测的研究背景;然后总结了已有软件缺陷数据属性度量方法的特点,并按照软件开发中缺陷预测的使用场景,以数据来源为主线从基于版本内数据、跨版本数据和跨项目数据实现缺陷预测三个方面对近10年（2005~2015）已有的研究工作进行分类归纳和比较;最后对该领域未来的研究趋势进行了展望.     

Performance of energy-aware and link-adaptive routing metrics for ultra wideband sensor networks

Wireless sensor networks are energy constrained since sensors operate with limited battery capacity. Thus, energy consumption is one of the most critical issues in the design of routing protocols. In addition, the link quality needs to be taken into account in the route decision for a wireless multihop network in order to efficiently exploit the inherent spatial diversity. In this paper we consider energy-aware and link-adaptive routing strategies for UWB (Ultra Wide Band) sensor networks. We utilize the ranging capabilities offered by UWB and employ adaptive modulation to take advantage of favorable link conditions. Different routing metrics are proposed based on the availability of sensor node’s location, link quality and next hop battery capacity information. These routing metrics integrate the measure of next hop remaining battery capacity with the throughput performance measures, Maximum Forward Progress (MFP) or Maximum Information Progress (MIP). The effectiveness of these metrics is evaluated in different simulation scenarios in terms of network throughput and lifetime for both random and grid sensor network topologies. Jinghao Xuis a D.Sc. candidate of the Department of Electrical and Computer Engineering at the George Washington University. He received his B.Sc. and M.S. degrees in telecommunication and electrical engineering from the Tianjin University, China, in 1993 and 1996, respectively. From 1996 to 2001, he was with Research Institute of Telecommunication Transmission of Ministry of Information Industry, Beijing, China, and China Wireless Telecommunication Standards group, where he was involved in the standardization of IMT-2000 communication systems for China. His research interests include performance evaluation and modeling of wireless ad hoc networks, Ultra Wide Band systems and multiuser detection techniques. Bojan M. Peric received the B.S. degree in electrical engineering from the University of Belgrade, Belgrade, Serbia and Montenegro in 2001. He is currently working toward the D.Sc. degree in electrical engineering at the George Washington University, Washington, DC. His research interests include wireless communications, with an emphasis on ad hoc networks. Branimir R. Vojcic is a professor in, and a past Chairman of, the Department of Electrical and Computer Engineering at the George Washington University. He has received his Dipl. Ing., M.Sc. and D.Sc. degrees from the University of Belgrade in Serbia and Montenegro in 1980, 1986 and 1989, respectively. His current research interests are in the areas of communication theory, performance evaluation and modeling mobile and wireless networks, mobile internet, code division multiple access, multiuser detection, adaptive antenna arrays, space-time coding and ad-hoc networks. He has also been an industry consultant and has published and lectured extensively in these areas. He co-authored the book: The cdma2000 System for Mobile Communications. Dr Vojcic received NSF CAREER Award in 1995. He is a Senior Member of IEEE, was an Associate Editor for IEEE Communications Letters and is presently an Associate Editor for Journal on Communications and Networks.     

Parametric Evaluation of Video Motion Tracking Data Sets

Video tracking is a complex problem because the environment, in which video motion needs to be tracked, is widely varied based on the application and poses several constraints on the design and performance of the tracking system. Current datasets that are used to evaluate and compare video motion tracking algorithms use a cumulative performance measure without thoroughly analyzing the effect of these different constraints imposed by the environment. But it needs to analyze these constraints as parameters. The objective of this paper is to identify these parameters and define quantitative measures for these parameters to compare video datasets for motion tracking.     

A relationship between software coverage metrics and reliability

Software engineers have tacitly assumed that there exists a link between coverage and reliability. An argument is presented which leads to an inverse logarithmic relationship which depends upon the structure of the software and on a parameter called the strength of the coverage metric. The relationship is developed by considering reliability as a function of an idealized coverage metric and then interpreting the result for a range of actual coverage metrics.