By Topic

Signal-Image Technologies and Internet-Based System, 2007. SITIS '07. Third International IEEE Conference on

Date 16-18 Dec. 2007

Filter Results

Displaying Results 1 - 25 of 153
  • [Front cover]

    Page(s): C1
    Save to Project icon | Request Permissions | PDF file iconPDF (285 KB)  
    Freely Available from IEEE
  • [Title page i]

    Page(s): i
    Save to Project icon | Request Permissions | PDF file iconPDF (160 KB)  
    Freely Available from IEEE
  • [Title page iii]

    Page(s): iii
    Save to Project icon | Request Permissions | PDF file iconPDF (194 KB)  
    Freely Available from IEEE
  • [Copyright notice]

    Page(s): iv
    Save to Project icon | Request Permissions | PDF file iconPDF (45 KB)  
    Freely Available from IEEE
  • Table of contents

    Page(s): v - xiv
    Save to Project icon | Request Permissions | PDF file iconPDF (334 KB)  
    Freely Available from IEEE
  • Foreword

    Page(s): xv
    Save to Project icon | Request Permissions | PDF file iconPDF (154 KB)  
    Freely Available from IEEE
  • Objectives

    Page(s): xvi
    Save to Project icon | Request Permissions | PDF file iconPDF (155 KB)  
    Freely Available from IEEE
  • Keynotes and Tutorials

    Page(s): xvii
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (157 KB)  

    Provides an abstract for each of the keynote presentations and a brief professional biography of each presenter. The complete presentations were not made available for publication as part of the conference proceedings. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Committees

    Page(s): xviii - xxvi
    Save to Project icon | Request Permissions | PDF file iconPDF (185 KB)  
    Freely Available from IEEE
  • Editorial messages

    Page(s): xxvii - xxx
    Save to Project icon | Request Permissions | PDF file iconPDF (172 KB)  
    Freely Available from IEEE
  • Improve Image Annotation by Combining Multiple Models

    Page(s): 3 - 9
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (372 KB) |  | HTML iconHTML  

    Automatic image annotation is a promising methodology for image retrieval. However most current annotation models are not yet sophisticated enough to produce high quality annotations. Given an image, some irrelevant keywords to image contents are produced, which are a primary obstacle to getting high-quality image retrieval. In this paper an approach is proposed to improve automatic image annotation two directions. One is to combine annotation keywords produced by underlying three classic image annotation models of translation model, continuous-space relevance model and multiple Bernoulli relevance models, hoping to increase the number of potential correctly annotated keywords. Another is to remove irrelevant keywords to image semantics based on semantic similarity calculation using WordNet. To verify the proposed hybrid annotation model, we carried out the experiments on the widely used Corel image data set, and the reported experimental results showed that the proposed approach improved image annotation to some extent. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Semantic Framework for Spatiotemporal Data Representation

    Page(s): 10 - 17
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (651 KB) |  | HTML iconHTML  

    Spatiotemporal databases have received much attention since more and more applications, such as environment management and land management, have shown urgent requirements on the management of spatiotemporal information. But different applications have different requirements on describing spatiotemporal objects and spatiotemporal changes, and there is no systematic foundation for the modeling of spatiotemporal data. In this paper, we first study the characteristics of spatiotemporal changes. And then we build a semantic framework for the representation of spatiotemporal data. Spatiotemporal changes are classified into six types. We use three elements to represent these spatiotemporal changes, which are called lifecycle, descriptor and transformation. Through the three elements, a fundamental framework to model spatiotemporal data is achieved. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Methodology for Low-Cost Image Annotation Based on Conceptual Modeling: A Biological Example

    Page(s): 18 - 25
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (644 KB) |  | HTML iconHTML  

    In the context of data-intensive biology applications the overall annotation costs are rather high since experts generally need to be involved. In some cases it is possible to automate the annotation process. In this paper we propose an approach that consists in building an image database application that enables domain technicians to make annotations without compromizing their semantical precision. Our proposal relies mainly on a domain ontology coupled with a conceptual data model. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Content-Based Shape Search of 3D CAD Models with Relevance Feedback

    Page(s): 26 - 32
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (617 KB) |  | HTML iconHTML  

    Engineers and designers often need to search out similar CAD models to make reference to and reuse the previous designs. For most of CAD models, both geometrical information and topological information are important for similarity assessment. We proposed two shape signatures to describe the geometry and the topology of a CAD model respectively. A relevance feedback mechanism is introduced to combine the two dissimilarity distances measured by the two signatures. Experiments are conducted to evaluate the performance of the proposed method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extraction of Ambiguous Sequential Patterns with Least Minimum Generalization from Mismatch Clusters

    Page(s): 35 - 42
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (539 KB) |  | HTML iconHTML  

    An ambiguous query in sequence databases returns a set of similar subsequences, called a mismatch cluster, to the user. The inherent problem is that it is difficult for users to identify the characteristics of very large similar subsequences in a mismatch cluster. In order to support user comprehension of mismatch clusters, it is important to extract a set of ambiguous sequence patterns with the least minimum generalization in the mismatch cluster. The extraction of the ambiguous sequential pattern set requires an enormous amount of computational time, since we have to discover generalized patterns with minimum covers for the mismatch cluster from candidate generalized patterns. The present paper is a proposal for an iterative refinement method to extract ambiguous sequence patterns with minimum cover for mismatch clusters selected from a sequence database. It includes a proposal to use the method with a domain segmentation method to achieve an efficient pattern extraction. Moreover, a prototype implementing the two proposed methods has been applied to three datasets included in PROSITE in order to evaluate their usefulness. The proposed methods resulted in a high capability to extract ambiguous sequential patterns from mismatch clusters that are provided by an ambiguous query in the sequence database. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed Frequent Closed Itemsets Mining

    Page(s): 43 - 50
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (473 KB) |  | HTML iconHTML  

    As many large organizations have multiple data sources and the scale of dataset becomes larger and larger, it is inevitable to carry out data mining in the distributed environment. In this paper, we address the problem of mining global frequent closed itemsets in distributed environment. A novel algorithm is proposed to obtain global frequent closed itemsets with exact frequency and it is shown that the algorithm can determine all the global frequent closed itemsets. A new data structure is developed to maintain the closed itemsets. Then an efficient implementation is provided based on the structure. Experimental results show that the proposed algorithm is effective. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Effective Site Customization Based on Web Semantics and Usage Mining

    Page(s): 51 - 59
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (773 KB) |  | HTML iconHTML  

    The explosive growth of online data and the diversity of goals that may be pursued over the Web have significantly increased the monetary value of the Web traffic. To tap into this accelerating market, Web site operators try to increase their traffic by customizing their sites to the needs of specific users. Web site customization involves two great challenges: the effective identification of the user interests and the encapsulation of those interests into the sites' presentation and content. In this paper, we study how we can effectively detect the user interests that are hidden behind navigational patterns and we introduce a novel recommendation mechanism that employs Web mining techniques for correlating the identified interests to the sitespsila semantic content, in order to customize them to specific users. Our experimental evaluation shows that the user interests can be accurately detected from their navigational behavior and that our recommendation mechanism, which uses the identified interests, yields significant improvements in the sites' usability. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Adaptive Privacy Preserving Data Mining Model under Distributed Environment

    Page(s): 60 - 68
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (592 KB) |  | HTML iconHTML  

    Privacy preserving becomes an important issue in the development progress of data mining techniques, especially in distributed data mining. Secure multiparty computation methods are proposed to protect the privacy in distributed environment, but shows low performance under massive nodes. This paper presents an adaptive privacy preserving data mining model based on data perturbation method to improve the efficiency while preserving the privacy. Security capability of basic data perturbation is firstly analyzed and an adaptive enhancement method is proposed according to the eigen value decomposition based attacks. A light-weight protocol with homomorphic technique is proposed to perform the perturbation process under distributed environments. The experiment results show that the model has high controllable security and shows more efficiency in large scale distribution environment comparing to secure multiparty related methods. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Passage Retrieval Using Graph Vertices Comparison

    Page(s): 71 - 76
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (492 KB) |  | HTML iconHTML  

    In this paper, we describe an Information retrieval Model based on graph comparison. It is inspired from previous work such as KleinbergpsilaHits and Blondel et al.psilas model. Unlike previous methods, our model considers different types of nodes: text nodes (elements to retrieve and query) and term nodes, so that the resulting graph is a bipartite graph. The results on passage retrieval task show that high precision is improved using this model. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Linguistic Analysis of Users' Queries: Towards an Adaptive Information Retrieval System

    Page(s): 77 - 84
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (373 KB) |  | HTML iconHTML  

    Most of information retrieval systems transform natural language users' queries into bags of words that are matched to documents, also represented as bags of words. Through such process, the richness of the query is lost. In this paper we show that linguistic features of a query are good indicators to predict systems failure to answer it. The experiments described here are based on 42 systems or system variants and 50 TREC topics that consist of a descriptive part expressed in natural language. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Topic Detection via Participation Using Markov Logic Network

    Page(s): 85 - 91
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1109 KB) |  | HTML iconHTML  

    The advent of Web 2.0 enables the proliferation of online communities in which tremendous number of Internet users contribute and share enormous information. Proper exploitation of community structure help retrieving useful information and better understanding of their features. We employ Markov Logic Network to explore topic tracking by finding clusters, which represents latent topics, best fitting a set of rules. Rather than using contents in investigating discussions of a community, the user participation is used because it is believed that topics can be somehow reflected by the preferences of participation. User participation is also easier to process than text. The clustering results show this approach can reveal latent topics of a community effectively. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Analysis of Constructed Categories for Textual Classification Using Fuzzy Similarity and Agglomerative Hierarchical Methods

    Page(s): 92 - 99
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (399 KB) |  | HTML iconHTML  

    Ambiguity is a challenge faced by systems that handle natural language. To assuage the issue of linguistic ambiguities found in text classification, this work proposes a text categorizer using the methodology of Fuzzy Similarity. The grouping algorithms Stars and Cliques are adopted in the Agglomerative Hierarchical method and they identify the groups of texts by specifying some time of relationship rule to create categories based on the similarity analysis of the textual terms. The proposal is that based on the methodology suggested, categories can be created from the analysis of the degree of similarity of the texts to be classified, without needing to determine the number of initial categories. The combination of techniques proposed in the categorizerpsilas phases brought satisfactory results, proving to be efficient in textual classification. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The Application of Wavelet & DCSK Watermarking in Multimedia Security

    Page(s): 103 - 108
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (707 KB) |  | HTML iconHTML  

    Nowadays digital Multimedia copyright protection is up against austere challenge. The watermarking and encryption used together is more and more important in digital multimedia security (DMS) application. So in this paper, a novel approximation zero-tree-wise wavelet watermarking with difference chaos shift key coding encryption method of DMS information security is present. Firstly the DMS information is got by proprietor or authority as initial watermarks. Then encrypt key is generated by DMpsilas data. To encrypt initial watermarks' information is to apply chaos sequence as final watermarks' information. Then the final watermark is embedded into the approximation zero-tree-wise wavelet coefficients of the image by digital chaos shift key (DCSK) coding method that makes watermarkspsila information more secure and secret. The Chaos key is generated automatically by image decomposed with LUT mapped. As a result, approximation zero-tree-wise wavelet with DCSK coding watermarking is more secure and secret characteristic in protecting DMS information. Furthermore blind-detection watermarking can be achieved, encrypt key can be regain by DM data. That method possesses research value. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • UMQL: A Unified Multimedia Query Language

    Page(s): 109 - 115
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (437 KB) |  | HTML iconHTML  

    We propose a unified multimedia query language called UMQL, based on a semi-structured data model. In contrast to previous multimedia query languages that are either designed only for one particular medium or not powerful enough on multimedia query ability, UMQL is a general and powerful multimedia language which can support effectively queries based on content, structure, spatial relationships and temporal relationships of multimedia data, and finally has been shown to be very expressive and quite suitable for multimedia information retrieval. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimization Algorithm and Data Security Problem in Distributed Information Systems

    Page(s): 116 - 120
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (345 KB) |  | HTML iconHTML  

    In this work we present a new rule-discovery method for distributed information system. DIS is the system that connects a number of information systems using network communication technology. This communication can be driven by request for knowledge needed to predict for maximal optimization which missing values can be replaced. In this work we recall the notion of a distributed information system to talk about handling semantic inconsistencies between sites. Semantic inconsistencies are due to different interpretations of attributes and their values on the concepts level among sites. Different interpretations can be also linked with a different way of treating null values among sites. Some attributes might be just hidden because of the security reason. In such case we have to be certain that the missing data can not be reconstructed from the available data by any known null value imputation method and that some information in IS can not be uncovered as well. Assuming that one attribute is hidden at one of the sites of DIS we will try to reconstruct this attribute. In this paper we will also show which values have to be hidden from users to guarantee that the hidden attribute can not be reconstructed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.