Showing papers by "Vincent Oria published in 2005"

Robust and fast similarity search for moving object trajectories

Abstract: An important consideration in similarity-based retrieval of moving object trajectories is the definition of a distance function. The existing distance functions are usually sensitive to noise, shifts and scaling of data that commonly occur due to sensor failures, errors in detection techniques, disturbance signals, and different sampling rates. Cleaning data to eliminate these is not always possible. In this paper, we introduce a novel distance function, Edit Distance on Real sequence (EDR) which is robust against these data imperfections. Analysis and comparison of EDR with other popular distance functions, such as Euclidean distance, Dynamic Time Warping (DTW), Edit distance with Real Penalty (ERP), and Longest Common Subsequences (LCSS), indicate that EDR is more robust than Euclidean distance, DTW and ERP, and it is on average 50% more accurate than LCSS. We also develop three pruning techniques to improve the retrieval efficiency of EDR and show that these techniques can be combined effectively in a search, increasing the pruning power significantly. The experimental results confirm the superior efficiency of the combined methods.

Using multi-scale histograms to answer pattern existence and shape match queries

Abstract: Similarity-based querying of time series data can be categorized as pattern existence queries and shape match queries. Pattern existence queries find the time series data with certain patterns while shape match queries look for the time series data that have similar movement shapes. Existing proposals address one of these or the other. In this paper, we propose multi-scale time series histograms that can be used to answer both types of queries, thus offering users more flexibility. Multiple histogram levels allow querying at various precision levels. Most importantly, the distanc es of time series histograms at lower scale are lower bounds of the distances at higher scale, which guarantees that no false dismissals will be introduced when a multi-step filter ing process is used in answering shape match queries. We further propose to use averages of time series histograms to reduce the dimensionality and avoid computing the distances of full time series histograms. The experimental results show that multi-scale histograms can effectively find the patterns in time series data and answer shape match queries, even when the data contain noise, time shifting and scaling, or amplitude shifting and scaling.

Multi-level index for global and partial content-based image retrieval

Abstract: This article presents a quadtree-based data structure for effective indexing of images. An image is represented by a multi-level feature vector, computed by a recursive decomposition of the image into four quadrants and stored as a full fixed-depth balanced quadtree. A node of the quadtree stores a feature vector of the corresponding image quadrant. A more general quadtree-based structure called QUIP-tree (QUadtree-based Index for image retrieval and Pattern search) is used to index the multi-level feature vectors of the images and their quadrants. A QUIP-tree node is an entry to a set of clusters that groups similar quadrants according to some pre-defined distances. The QUIP-tree allows a multi-level filtering in content-based image retrieval as well as partial queries on images.

Evaluation of queries on tree-structured data using dimension graphs

Abstract: The recent proliferation of XML-based standards and technologies for managing data on the Web demonstrates the need for effective and efficient management of tree-structured data. Querying tree-structured data is a challenging issue due to the diversity of the structural aspect in the same or in different trees. In this paper, we show how to evaluate queries on tree-structured data, called value trees. The formulation of these queries does not depend on the structure of a particular value tree. Our approach exploits semantic information provided by dimension graphs. Dimension graphs are semantically rich constructs that abstract the structural information of the value trees. We show how dimension graphs can be used to query efficiently value trees in the presence of structural differences and irregularities. Value trees and their dimension graphs are represented as XML documents. We present a method for transforming queries to XPath expressions to be evaluated on the XML documents. We also provide conditions for identifying strongly and weakly unsatisfiable queries. Finally, we conducted various experiments to compare our method for evaluating queries with one that does not exploit dimension graphs. Our results demonstrate the superiority of our approach.

Report from the first international workshop on computer vision meets databases (CVDB 2004)

Abstract: This report summarizes the presentations and discussions of the First International Workshop on Computer Vision meets Databases, or CVDB 2004, which was held in Paris, France, on June 13, 2004. The workshop was co-located with the 2004 ACM SIGMOD/PODS conferences and was attended by forty-two participants from all over the world.