一个有效的基于运动的视频检索系统

在视频摄取的过程中,摄象机的运动在某种程度上反映了当前视频的部分语义特征,所以通过对摄象机运动的提取能够更好地实现对视频的浏览以及基于内容的检索。针对以上应用,本文提出了一种用运动矢量的分布来提取摄相机运动的方法,测试后表明,该方法能有效地分析出多种相机运动并具有较好的鲁棒性。同时我们以此为基础实现了一个基于相机运动的视频检索系统。     

基于XML的图形技术在WebGIS中的应用

空间信息技术和网络技术的融合,产生了一个基于Internet的WebGIS研究的热门课题。介绍了互联网信息时代的WebGIS的几种实现方案,认为传统的HTML在Web浏览器和服务器之间以图象格式传送JPG／GIF的地图数据,限制了WebGIS的性能,提出了一种基于可升级矢量图形SVG（Scalable Vector Graphics)和组件技术的WebGIS的解决方案。     

图像集的离散度矩阵及其在人脸图像识别中的应用

本文提出了图像集离散度矩阵的基本概念及用于人脸图像识别的方法。该方法把同一类图像集中的各幅图像的信息矩阵的奇异值向量作为矩阵中的一列构成图像集特征矩阵。把这些矩阵组成一图像库,并找出图像集离散度矩阵和图像库离散度矩阵,从而计算出它们的最佳投影系数。然后,把测试样本的图像集特征矩阵与图像集库中的训练样本图像集的特征按投影系数相比较找出它的相似程度,从而进行人脸图像识别。     

一种利用多传感器进行目标识别的方法

针对传统单传感器目标识别利用目标信息有限、识别结果可靠性不高等缺陷 ,提出一种利用多传感器进行目标识别方法 ,这种方法充分利用了目标多重特征的信息 ,并考虑了不同特征信息在目标识别中的重要程度 ,从而得到更为接近实际效果的识别结果     

Constraint—Preserving Architecture Transformations：A Graph Rewriting Approach

Architecture transformations are frequently performed during software design and maintenance.However this activity is not well supported at a sufficiently abstract level.In this paper,the authors characterize architecture transformations using using graph rewriting rules,where architectures are represented in graph notations,Architectures are usually required to satisfy certain constraints during evolution.Therefore a way is presented to construct the sufficient and necessary condition for a transformatio to preserve a constraint.The condition can be verified before the application of the transformation.Validated transformations are guaranteed not to violate corresponding constraints whenever applied.     

基于简化系统依赖图的静态粗粒度切片方法

基于系统依赖图是计算面向对象程序切片的一个有效方法.但是,系统依赖图的缺点是太复杂,而且在建立系统依赖图的过程中容易出错,一旦出现错误就可能导致切片结果的不准确.通过对系统依赖图进行简化,得到了简化的系统依赖图.它省略了那些表示输入参数和输出参数的结点和概括边.同时,还定义了一种面向对象程序的粗粒度切片概念,讨论了它的性质,分析了它与细粒度切片的关系,并基于简化的系统依赖图计算面向对象程序的粗粒度切片.最后还讨论了切片技术的简单实现.     

基于最小连通邻域图的ISOMAP算法

噪音的干扰和邻域大小的不合适会在ISOMAP算法的邻域图中引入“短路”边,使其不能正确表达数据的邻域结构,从而使该算法具有较差的鲁棒性和拓扑稳定性。为此,根据最小连通邻域图能有效避免“短路”边的特点,提出了一种能有效删除“短路”边因而更具鲁棒性和拓扑稳定性的ISOMAP算法——基于最小连通邻域图的ISOMAP（MCNG-ISOMAP）算法。该算法能在一定程度上避免邻域大小难以有效选取的问题,同时还能在不依赖于邻域大小的情况下发现数据真正的固有维数。     

Sparse least squares support vector training in the reduced empirical feature space

In this paper we discuss sparse least squares support vector machines (sparse LS SVMs) trained in the empirical feature space, which is spanned by the mapped training data. First, we show that the kernel associated with the empirical feature space gives the same value with that of the kernel associated with the feature space if one of the arguments of the kernels is mapped into the empirical feature space by the mapping function associated with the feature space. Using this fact, we show that training and testing of kernel-based methods can be done in the empirical feature space and that training of LS SVMs in the empirical feature space results in solving a set of linear equations. We then derive the sparse LS SVMs restricting the linearly independent training data in the empirical feature space by the Cholesky factorization. Support vectors correspond to the selected training data and they do not change even if the value of the margin parameter is changed. Thus for linear kernels, the number of support vectors is the number of input variables at most. By computer experiments we show that we can reduce the number of support vectors without deteriorating the generalization ability.

Modelling Calculi with Name Mobility using Graphs with Equivalences

In the theory of graph rewriting, the use of coalescing rules, i.e., of rules which besides deleting and generating graph items, can coalesce some parts of the graph, turns out to be quite useful for modelling purposes, but, at the same time, problematic for the development of a satisfactory partial order concurrent semantics for rewrites. Rewriting over graphs with equivalences, i.e., (typed hyper)-graphs equipped with an equivalence over nodes provides a technically convenient replacement of graph rewriting with coalescing rules, for which a truly concurrent semantics can be easily defined. The expressivity of such a formalism is tested in a setting where coalescing rules typically play a basic role: the encoding of calculi with name passing as graph rewriting systems. Specifically, we show how the (monadic fragment) of the solo calculus, one of the dialect of those calculi whose distinctive feature is name fusion, can be encoded as a rewriting system over graph with equivalences.     

A low-power wireless video sensor node for distributed object detection

In this paper we propose MicrelEye, a wireless video node for cooperative distributed video processing applications that involve image classification. The node is equipped with a low-cost VGA CMOS image sensor, a reconfigurable processing engine (FPGA, Microcontroller, SRAM) and a Bluetooth 100-m transceiver. It has a size of few cubic centimeters and its typical power consumption is approximately ten times less than that of typical commercial DSP-based solutions. As regards classification, a highly optimized hardware-oriented support vector machine-like (SVM-like) algorithm called ERSVM is proposed and implemented. We describe our hardware and software architecture, its performance and power characteristics. The case study considered in this paper is people detection. The obtained results suggest that the present technology allows for the design of simple intelligent video nodes capable of performing classification tasks locally.