以太网交换机的源指定组播方案

针对边缘网中组播数据转发效率不高的问题,提出以太网交换机的源指定组播方案。在该方案中,交换机利用对IGMPv3报文的嗅探实现指定源的组播数据转发。描述控制流和数据流的运行机制,对该机制进行仿真分析。与IGMPv2嗅探方案的比较结果证明,该方案性能较高。     

基于CNN的监控视频事件检测

复杂监控视频中事件检测是一个具有挑战性的难题, 而TRECVID-SED评测使用的数据集取自机场的实际监控视频,以高难度著称. 针对TRECVID-SED评测集, 提出了一种基于卷积神经网络(Convolutional neural network, CNN)级联网络和轨迹分析的监控视频事件检测综合方案. 在该方案中, 引入级联CNN网络在拥挤场景中准确地检测行人, 为跟踪行人奠定了基础; 采用CNN网络检测具有关键姿态的个体事件, 引入轨迹分析方法检测群体事件. 该方案在国际评测中取得了很好的评测排名: 在6个事件检测的评测中, 3个事件检测排名第一.     

网络空间终端设备识别框架

针对传统Web服务识别方法对服务器软件实现差异依赖性强、不利于扩展到其他终端设备等缺陷,提出了一种网络空间终端设备双重因素主动识别框架. 该框架从服务标识（Banner）和Web指纹两个角度分析终端设备网络指纹特征. 首先,从HTTP数据包头部的“Server”字段、“WWW-Authenticate”字段及HTML源代码中深入挖掘终端设备Web服务中的Banner信息;其次,从图片文件和文本文件中提取终端设备具有的Web指纹特征;同时,结合具体案例分析各识别方法的有效性及优缺点. 结果表明,该框架解决了传统识别方法中Banner信息获取途径单一、依赖服务器软件差异特征的问题,成功将Web服务识别方法扩展到了网络空间终端设备上. 最后,总结已有成果并探讨该框架未来改进方向.     

面向视频压缩的显著性协同检测JND模型

为了更好的将人眼感知特性用于视频压缩系统,提出了一种改进的基于显著性协同检测的恰可察觉失真模型（Just Noticeable Distortion,JND）.该模型通过像素域和变换域下联合建模计算得到的最优JND模型,基于上下文感知的显著性算法得到相应的显著图,并将检测结果用于JND模型权值分配.提出的JND残差滤波器可以嵌入到HEVC视频编码框架中.实验结果表明:在全I帧配置下,提出的算法编码结果与HM16相比,在视觉主观感知质量一致的情况下,平均码率可节省10.7%.     

Image contrast enhancement using fuzzy clustering with adaptive cluster parameter and sub-histogram equalization

Histogram equalization is an effective technique to boost image quality and contrast enhancement. However, in some cases the increase in image contrast by traditional histogram equalization exceeds the desired amount Which damages the image properties and wanes its natural look. Histogram division and performing a separate equalization for each sub-histogram is one of the presented solutions. The dividing method and determining the number of sub-histograms are the main problems directly affecting the output image quality. In this study, a method is introduced for automatic determination of the number of sub-histograms and density based histogram division leading to appropriate output with no need for parameter setting. Each main peak is in a separate section. Image contrast is increased with no loss of image specifications through determining the number of sub-histograms based on the number of main peaks. The introduced histogram equalization approach consists of three stages. The first stage, using histogram analysis, produces an automated estimate of number of clusters for image brightness levels. The second, clusters the image brightness levels, and using the provided transfer function, the final stage includes contrast enhancement for each individual cluster separately. The results of the proposed approach demonstrate not only clearer details along with a boost in contrast, but also noticeably more natural appearance in the images.     

A feature dependent Naive Bayes approach and its application to the software defect prediction problem

Naive Bayes is one of the most widely used algorithms in classification problems because of its simplicity, effectiveness, and robustness. It is suitable for many learning scenarios, such as image classification, fraud detection, web mining, and text classification. Naive Bayes is a probabilistic approach based on assumptions that features are independent of each other and that their weights are equally important. However, in practice, features may be interrelated. In that case, such assumptions may cause a dramatic decrease in performance. In this study, by following preprocessing steps, a Feature Dependent Naive Bayes (FDNB) classification method is proposed. Features are included for calculation as pairs to create dependence between one another. This method was applied to the software defect prediction problem and experiments were carried out using widely recognized NASA PROMISE data sets. The obtained results show that this new method is more successful than the standard Naive Bayes approach and that it has a competitive performance with other feature-weighting techniques. A further aim of this study is to demonstrate that to be reliable, a learning model must be constructed by using only training data, as otherwise misleading results arise from the use of the entire data set.     

基于Spark的BIRCH算法并行化的设计与实现

在分布式计算和内存为王的时代,Spark作为基于内存计算的分布式框架技术得到了前所未有的关注与应用。着重研究BIRCH算法在Spark上并行化的设计和实现,经过理论性能分析得到并行化过程中时间消耗较多的Spark转化操作,同时根据并行化BIRCH算法的有向无环图DAG,减少shuffle和磁盘读写频率,以期达到性能优化。最后,将并行化后的BIRCH算法分别与单机的BIRCH算法和MLlib中的K-Means聚类算法做了性能对比实验。实验结果表明,通过Spark对BIRCH算法并行化,其聚类质量没有明显的损失,并且获得了比较理想的运行时间和加速比。     

基于安卓平台的电力作业信息系统的设计

文章实现了电力作业信息系统的设计,包括电力作业软件的设计以及Webservice服务器的组成结构。文章着重描述了电力作业手机软件GUI的设计、电子地图开发、数据库的设计方法;详细介绍了服务器端WebService的搭建及服务发布流程、服务器接口的设计。系统服务器采用Linux、Oracle数据库及Axis2＋Tomcat搭建WebService平台,采用第三方类库提供轻量级SOAP客户端的Ksoap2实现服务器与Android客户端数据的交互。本系统已经在义乌电力局进行试用,系统性能稳定、兼容性好、具有良好的可扩展性和广泛的应用价值。     

基于OpenF1oW的可编程终端设计与实现

随着互联网和网络接入技术的不断发展,终端配备多个网络接口已经十分普遍,多个接口同时接入网络,会建立多条连接,现有的多接口终端除了修改内核之外,没有一种灵活的方法管理多个连接。OpenFlow技术已经吸gj了越来越多网络研究人员的关注,本文中将OpenFlow技术应用到终端,兼容现有的网络协议栈的同时,使终端具有可编程性,并提供了一种在可编程终端上管理多连接的方法,基于终端控制器开发实现了连接管理组件,并且在Linux系统上成功实现。