Similarity-based image organization and browsing using multi-resolution self-organizing map

An algorithm is presented in this paper to facilitate the exploration of large image collections based on visual similarities. Starting with an unordered and unannotated set of images, the algorithm first extracts the salient details into feature vectors using both color and gradient information. The feature vectors are then used to train a self-organizing map which maps high-dimensional feature vectors onto a 2D canvas so that images with similar feature vectors are grouped together. When users browse the image collection, an image collage is generated that selects and displays the most pertinent set of images based on which portion of the 2D canvas is currently in view. Flowing from an overview to details is a seamless operation controlled simply by pan and zoom, with representative images selected in a consistent and predictable way. To make organizing larger image collections practical in interactive time, the organization algorithm is designed to run in parallel on graphics processing units. Overall this paper presents an end-to-end solution that facilitates the surfing of image collections in a fresh way.     

基于智能流量预测的入侵检测方法

为了提高入侵检测系统检测的实时性,提出了一种基于智能流量预测的入侵检测方法。该方法拟合了智能Agent的智能性、自主性和自适应性的优点以及灰色预测对不确定资源的科学预测的优点,用流量预测智能Agent预测得到的流量序列来代替未来一段时间段内的实际流量,并把这个预测序列作为检测对象集的一部分。然后用人工方法模拟了流量处理Agent与预测智能Agent的活动,并通过对实际的采集数据进行仿真,实验证明了预测智能Agent的预测活动的科学性。     

基于Gabor滤波器的指纹图像增强算法

介绍了一种基于Gabor滤波器的指纹图像增强算法,该方法对传统的Gabor滤波器的参数和大小进行了优化。实验表明这种算法具有很好的处理效果。     

异地并行设计中WEB数据库设计

分析了基于Internet的产品异地并行设计系统中的数据流特点,对用于产品异地并行设计的Web数据库结构、与Internet的接口技术以及确保数据安全的技术方法进行了较为深入的研究。     

GIS和GPS在交通中的应用

地理信息系统（GIS）正在以前所未有的速度应用到各个领域中,而全球定位系统（GPS）也已开始应用于交通运输和道路工程之中,该文介绍了利用GIS软件实现车载GPS实时跟踪并在指挥中心的数字地图上集中显示。     

A TLP approach for BGP based on local speculation

With the explosive growth of Internet, the low efficiency of BGP has seriously influenced its usability. In this work, a TLP （thread level parallelism） approach with local speculation is proposed to improve the BGP performance. The thread partition is locally performed on each separated sub-module at route processing, and the speculation strategy is implemented to guarantee the memory consistency and sequential commit. Experiments on Intel Quad-core server show that this approach reaches an average speedup of 1.46 under single peer, multi-peers and route flapping. It is also shown that the packet throughput can be improved greatly under multiple sessions by taking advantage of TLP.     

Direct-Referencing Two-Dimensional-Array Digital Microfluidics Using Multilayer Printed Circuit Board

Digital (i.e., droplet-based) microfluidics, by the electrowetting-on-dielectric (EWOD) mechanism, has shown great potential for a wide range of applications, such as lab-on-a-chip. While most reported EWOD chips use a series of electrode pads essentially in 1D line pattern designed for specific tasks, the desired universal chips allowing user-reconfigurable paths would require the electrode pads in 2D pattern. However, to electrically access the electrode pads independently, conductive lines need to be fabricated underneath the pads in multiple layers, raising a cost issue particularly for disposable chip applications. In this paper, we report the building of digital microfluidic plates based on a printed circuit board (PCB), in which multilayer electrical access lines were created inexpensively using the mature PCB technology. However, due to its surface topography and roughness and resulting high resistance against droplet movement, the as-fabricated PCB surfaces require high (~500 V) voltages unless coated with or immersed in oil. Our goal is the EWOD operations of droplets not only on oil-covered surfaces but also on dry ones. To meet the varying levels of performances, three types of gradually complex post-PCB microfabrication process are developed and evaluated. By introducing land-grid-array sockets in the packaging, a scalable digital microfluidic system with a reconfigurable and low-cost chip is also demonstrated.     

关联规则在购物篮分析中的应用

购物篮分析是数据挖掘技术应用在零售业中的一种有效方式,其目的就是在顾客的购买交易中分析出能够同时购买一类产品或一组产品的可能性,利于商品的摆放,也利于提高促销活动的效果.以一个面向小型超市的购物篮分析应用实例为背景,利用Apriori算法找出满足最小支持度阈值和最小置信度阈值的强关联规则, 并运用此强关联规则给超市物品摆放提供建议性的意见.     

基于DOS的仿真技术在Windows环境下的应用

介绍如何运用可视化技术,对已有基于DOS的应用系统进行可视化仿真,并对在Windows事件驱动环境下,如何解决流程控制应用系统仿真中的逻辑难于控制的问题进行了深入讨论,最后提出了逻辑集中控制的仿真策略。     

Efficient Nonnegative Matrix Factorization via projected Newton method

Nonnegative Matrix Factorization (NMF) is a popular decomposition technique in pattern analysis, document clustering, image processing and related fields. In this paper, we propose a fast NMF algorithm via Projected Newton Method (PNM). First, we propose PNM to efficiently solve a nonnegative least squares problem, which achieves a quadratic convergence rate under appropriate assumptions. Second, in the framework of an alternating optimization method, we adopt PNM as an essential subroutine to efficiently solve the NMF problem. Moreover, by exploiting the low rank assumption of NMF, we make PNM very suitable for solving NMF efficiently. Empirical studies on both synthetic and real-world (text and image) data demonstrate that PNM is quite efficient to solve NMF compared with several state of the art algorithms.