一种基于多时间粒度的数据流建模方法

在很多领域中,越来越多的数据以数据流的形式存在于各种应用当中,这些数据的特点是实时的、连续的、时变的、快速的。由于这些特点,在数据处理方法上引入了很多挑战性问题。本文重点从多时间粒度的角度研究了数据流建模问题,提出了多时间粒度的数据流滑窗建模方法,采用层次窗口模型对数据流进行描述,有效解决了Ad-Hoc查询
中的历史数据管理问题。     

Oporto: A Realistic Scenario Generator for Moving Objects

The spatio-temporal database research community has just started to investigate benchmarking issues. On one hand we would rather have a benchmark that is representative of real world applications, in order to verify the expressiveness of proposed models. On the other hand, we would like a benchmark that offers a sizeable workload of data and query sets, which could obviously stress the strengths and weaknesses of a broad range of data access methods. This paper offers a framework for a spatio-temporal data sets generator, a first step towards a full benchmark for the large real world application field of smoothly moving objects with few or no restrictions in motion. The driving application is the modeling of fishing ships where the ships go in the direction of the most attractive shoals of fish while trying to avoid storm areas. Shoals are themselves attracted by plankton areas. Ships are moving points; plankton or storm areas are regions with fixed center but moving shape; and shoals are moving regions. The specification is written in such a way that the users can easily adjust generation model parameters.     

Tracking Multiple Moving Objects for Real-Time Robot Navigation

This paper proposes a method for detecting and tracking the motion of a large number of dynamic objects in crowded environments, such as concourses in railway stations or airports, shopping malls, or convention centers. With this motion information, a mobile vehicle is able to navigate autonomously among moving obstacles, operating at higher speeds and using more informed locomotion strategies that perform better than simple reactive manoeuvering strategies. Unlike many of the methods for motion detection and tracking discussed in the literature, our approach is not based on visual imagery but uses 2D range data obtained using a laser rangefinder. The direct availability of range information contributes to the real-time performance of our approach, which is a primary goal of the project, since the purpose of the vehicle is the transport of humans in crowded areas. Motion detection and tracking of dynamic objects is done by constructing a sequence of temporal lattice maps. These capture the time-varying nature of the environment, and are denoted as time-stamp maps. A time-stamp map is a projection of range information obtained over a short interval of time (a scan) onto a two-dimensional grid, where each cell which coincides with a specific range value is assigned a time stamp. Based on this representation, we devised two algorithms for motion detection and motion tracking. The approach is very efficient, with a complete cycle involving both motion detection and tracking taking 6 ms on a Pentium 166 MHz. The system has been demonstrated on an intelligent wheelchair operating in railway stations and convention centers during rush hour.     

Indexing the Trajectories of Moving Objects in Networks*

The management of moving objects has been intensively studied in recent years. A wide and increasing range of database applications has to deal with spatial objects whose position changes continuously over time, called moving objects. The main interest of these applications is to efficiently store and query the positions of these continuously moving objects. To achieve this goal, index structures are required. The main proposals of index structures for moving objects deal with unconstrained 2-dimensional movement. Constrained movement is a special and a very important case of object movement. For example, cars move in roads and trains in railroads. In this paper we propose a new index structure for moving objects on networks, the MON-Tree. We describe two network models that can be indexed by the MON-Tree. The first model is edge oriented, i.e., the network consists of nodes and edges and there is a polyline associated to each edge. The second one is more suitable for transportation networks and is route oriented, i.e., the network consists of routes and junctions. In this model, a polyline also serves as a representation of the routes. We propose the index in terms of the basic algorithms for insertion and querying. We test our proposal in an extensive experimental evaluation with generated data sets using as underlying networks the roads of Germany. In our tests, the MON-Tree shows good scalabiliy and outperforms the competing index structures in updating (index creation) as well as in querying.*This work was partially supported by a grant Gu 293/8–l from the Deutsche Forschungsgemeinschaft (DFG), project Datenbanken für bewegte Objekte (Databases for Moving Objects).     

基于受限移动对象当前及将来时刻检索的时空索引结构研究

在给定的空间及时间范围内,如何构建高效的时空索引结构,以实现对移动对象快速有效的检索,是实现定位服务、智能交通、数字化战争等诸多应用中所迫切需要解决的问题.本文依据移动对象的运动特点,提出了一种面向当前及将来时刻快速更新及有效检索的索引结构—PQR树.PQR树是综合PMRQuad树和R*树的结构,首先依据道路分布用PMRQuad树将移动对象的索引空间实行粗略的层分割,将所有快速移动对象与道路相关联.然后用R*树索引分布在各个子空间块内的类静止对象.实验结果表明PQR树具有良好的更新和查询性能.     

Spatio-Temporal Data Types: An Approach to Modeling and Querying Moving Objects in Databases

Spatio-temporal databases deal with geometries changing over time. In general, geometries cannot only change in discrete steps, but continuously, and we are talking about moving objects. If only the position in space of an object is relevant, then moving point is a basic abstraction; if also the extent is of interest, then the moving region abstraction captures moving as well as growing or shrinking regions. We propose a new line of research where moving points and moving regions are viewed as 3-D (2-D space+time) or higher-dimensional entities whose structure and behavior is captured by modeling them as abstract data types. Such types can be integrated as base (attribute) data types into relational, object-oriented, or other DBMS data models; they can be implemented as data blades, cartridges, etc. for extensible DBMSs. We expect these spatio-temporal data types to play a similarly fundamental role for spatio-temporal databases as spatial data types have played for spatial databases. The paper explains the approach and discusses several fundamental issues and questions related to it that need to be clarified before delving into specific designs of spatio- temporal algebras.     

Modeling and Retrieval of Moving Objects

This paper presents a symbolic formalism for modeling and retrieving video data via the moving objects contained in the video images. The model integrates the representations of individual moving objects in a scene with the time-varying relationships between them by incorporating both the notions of object tracks and temporal sequences of PIRs (projection interval relationships). The model is supported by a set of operations which form the basis of a moving object algebra. This algebra allows one to retrieve scenes and information from scenes by specifying both spatial and temporal properties of the objects involved. It also provides operations to create new scenes from existing ones. A prototype implementation is described which allows queries to be specified either via an animation sketch or using the moving object algebra.     

面向形态丰富语言的多粒度翻译融合

形态丰富语言由于其复杂的形态变化,会导致大词汇量和数据稀疏问题,这给统计机器翻译带来了巨大挑战。该文通过将这类语言表示为不同的粒度,然后分别进行翻译;由于不同的粒度能表征语言不同层面的特点,通过对不同粒度的翻译结果进行词级系统融合,便可生成更好的译文。维吾尔语、蒙古语到汉语的两组翻译实验表明,这种多粒度系统融合方法改善了翻译效果,BLEU值比最好的单系统分别提高了+1.41%和+2.03%。     

基于水平集的多运动目标时空分割与跟踪

针对背景运动时的运动目标分割问题,提出了一种对视频序列中的多个运动目标进行分割和跟踪的新方法。该方法着眼于运动的且较为复杂的背景,首先利用光流约束方程和背景运动模型建立一个基于时空域的能量函数,然后用该函数进行背景运动速度的估算和运动目标的分割和跟踪。而时空域中的运动目标的最佳分割,乃是通过使该能量函数最小化来驱动时空曲面演化实现。时空曲面的演化采用了水平集PDEs(Partial Differential Equations)方法。实验中,用实际的图像序列验证了该算法及其数值实现。实验表明,该方法能够同时进行背景运动速度的估算、运动目标的分割和跟踪。     

不确定移动对象的模糊时空范围查询

指出不确定性和模糊性在时空语义上的区别;提出不确定移动对象的模糊时空范围查询问题,即查询条件中时间、空间范围的外延是模糊的,无清晰的边界,而目标对象的位置不确定;用模糊集表示模糊查询条件,概率密度函数表示移动对象在各自不确定区域内的可能位置分布;给出了不确定对象关于模糊查询条件匹配度的计算方法;设计了基于α截集的无效对象排除和有效对象确认规则及查询算法.算法规则适用于任意概率密度分布.现有的确定或不确定范围查询可以看成是模糊时空范围查询的特例.通过实验验证了算法的效率,在各种参数设置下,约有30%~90%的查询结果可在不计算匹配度的情况下获得.     

时间区间上的不确定移动对象距离范围查询

针对目标对象与查询发出者皆为不确定移动对象的情况,提出了一种时间区间上的距离范围查询（DRqTI）.此类查询搜索出数据集中在给定时间区间内,到查询发出者距离不超过阈值的目标对象,查询结果中包含对象满足查询条件的有效时间段和匹配度.提出了基于轨迹、基于时间区间和基于距离的三种剪枝策略,并给出了精炼和匹配度计算方法,在此基础上设计了查询处理算法.实验分析表明,三种剪枝策略中基于距离的方法性能最佳,提出的算法能有效处理DRqTI问题.     

时空数据库移动点轨迹建模

时空数据库的移动点轨迹建模技术是近年来是时空数据库研究热点之一.本文在给定一系列时空数据库移动点的信息,找出实际应用中时间更新阈值和空间距离阈值,结合移动点时空模型和数学上线性回归的建模思想,建立一种新的支持过去、现在、未来的移动点轨迹模型,并有效地预测了移动点在某一时间阈值内,移动点的运动趋势及范围.     

一种基于多时间粒度的实时客流查询优化算法

随时间实时变化的客流数据属于时间序列数据,本文根据客流数据的接收频率,应用关系模型实现客流数据的存储建模;为了减弱数据采集频率对实时客流查询效率的影响,建立多时间粒度的客流视图,可提高实时客流查询的计算效率.     

Reporting Leaders and Followers among Trajectories of Moving Point Objects

Widespread availability of location aware devices (such as GPS receivers) promotes capture of detailed movement trajectories of people, animals, vehicles and other moving objects, opening new options for a better understanding of the processes involved. In this paper we investigate spatio-temporal movement patterns in large tracking data sets. We present a natural definition of the pattern ‘one object is leading others’, which is based on behavioural patterns discussed in the behavioural ecology literature. Such leadership patterns can be characterised by a minimum time length for which they have to exist and by a minimum number of entities involved in the pattern. Furthermore, we distinguish two models (discrete and continuous) of the time axis for which patterns can start and end. For all variants of these leadership patterns, we describe algorithms for their detection, given the trajectories of a group of moving entities. A theoretical analysis as well as experiments show that these algorithms efficiently report leadership patterns.

复杂背景中一种特定运动目标检测与跟踪方法

针对复杂环境对运动目标检测与跟踪产生的不利影响,提出一种自适应运动能量阈值结合精简彩色SIFT描述子的特定运动目标检测与跟踪方法。运用自适应运动能量阈值方法自动滤除复杂环境干扰以完成运动目标检测,由此形成目标匹配搜索域,并给出经主成份分析及精简后的彩色SIFT描述子( PCA-CSIFT )进行目标匹配,从而实现特定运动目标的连续跟踪。实验结果表明,在复杂环境下,运动目标检测方法对目标总量变化不敏感,错误率始终稳定在6.5%~34%之间。 PCA-CSIFT算法在保持高可区分性的同时错误匹配率为25.33%~28%,平均每帧处理时间不超过0.26 s,具有较好的鲁棒性与实时性。     

基于Petri网的移动对象的时空约束

本文利用Petri网技术提出了移动对象的时空约束关系。文中首先利用V41理论给出了移动对象的空间约束,之后将该理论应用于时态中给出了移动对象的时态约束。在提出的移动对象Petri网(MOPN)和空间约束Petri网(SCPlN)的基础上,给出了移动对象的时空约束关系。     

RFID数据流上多目标复杂事件检测

已有的RFID复杂事件处理技术主要关注于单个RFID对象的复杂事件检测和优化技术.实际上,很多RFID应用中往往需要同时检测多个同类型关联目标的复杂事件序列.研究了多个关联的RFID对象的复杂事件处理问题.通过扩展的事件语言和算子的语义以支持同类型多个RFID目标复杂事件查询的定义.通过模式的变换规则,将RFID应用中存在的各种非线性多目标复杂事件模式转换成线性模式,以便各种多目标模式在一个统一的框架下检测.提出了基于自动机NFA\\-{b2}的多目标复杂事件检测模型和多目标复杂事件检测算法.通过在多目标检测算法中使用关键节点下压和同位置约束置后优化策略,大大减少了单个类型上无用实例的数目和不同类型间模式匹配的搜索空间.与SASE算法的实验比较表明算法的正确性和高效性.     

位置服务中对移动物体的时空索引

在TPR*-tree的基础上提出了DTPR*-tree索引结构, 引进了事务时间和有效时间.它能索引移动物体过去、当前和未来的位置信息,支持基本位置查询和空间约束查询,有很好的空间、查询和更新效率.     

移动对象数据库模型、查询语言及实时交通流分析

提出一种移动对象数据库模型——Dynamic Transportation Network Based Moving Objects Database(简称DTNMOD),并给出了DTNMOD中基于移动对象时空轨迹的网络实时动态交通流分析方法.在DTNMOD中,交通网络被表示成动态的时空网络,可以描述交通状态、拓扑结构以及交通参数随时间的变化过程;网络受限的移动对象则用网络移动点表示.DTNMOD 模型包含了完整的数据类型和查询操作的定义,因此可以在任何可扩充数据库(如PostgreSQL 或SECONDO)中实现,从而得到完整的数据库模型和查询语言.为了对相关模型的性能进行比较与分析,基于PostgreSQL 实现了一个原型系统并进行了一系列的实验.实验结果表明,DTNMOD 提供了良好的区域查询及连接查询性能.     

基于区域活动轮廓运动目标跟踪方法研究

根据贝叶斯估计理论，首先建立了图像序列中运动目标的跟踪模型，然后用高斯分布来描述图像的区域信息，并通过对模型的分析，与区域活动轮廓模型建立对应关系，将问题的求解转化为能量最小化问题。同时为了克服目标在运动中发生的拓扑形变，采用水平集方法进行数值实现。实验结果表明，这种方法不仅可以对多个运动目标进行跟踪，并能非常好地逼近运动目标的轮廓，而且能够自然地处理运动目标的拓扑形变。