Personal tools
You are here: Home Publications Technical Reports
Document Actions

Technical Reports

Up one level
Evaluating the Potential of Texture and Color Descriptors for Remote Sensing Image Retrieval and Classification
Classifying Remote Sensing Images (RSI) is a hard task. There are automatic approaches whose results normally need to be revised. The identification and polygon extraction tasks usually rely on applying classification strategies that exploit visual aspects related to spectral and texture patterns identified in RSI regions. There are a lot of image descriptors proposed in the literature for content-based image retrieval purposes that can be useful for RSI classification. This paper presents a comparative study to evaluate the potential of using successful color and texture image descriptors for remote sensing retrieval and classification. Seven descriptors that encode texture information and twelve color descriptors that can be used to encode spectral information were selected. We highlight the main characteristics and perform experiments to evaluate the effectiveness of these descriptors. To evaluate descriptors in classification tasks, we also proposed a methodology based on KNN classifier. Experiments demonstrate that Joint Auto-Correlogram (JAC), Color Bitmap, Invariant Steerable Pyramid Decomposition (SID) and Quantized Compound Change Histogram (QCCH) yield the best results.
SPICA's Multi-party Negotiation Protocol: Implementation using YAWL
A supply chain comprises several different kind of actors that interact either in an ad hoc fashion (e.g., an eventual deal) or in a previously well planned way. In the latter case, how the interactions develop is described in contracts that are agreed on before the interactions start. This agreement may involve several partners, thus a multi-party contract is better suited than a set of bi-lateral contracts. If one is willing to negotiate automatically such kind of contracts, an appropriate negotiation protocol should be at hand. However, the ones for bi-lateral contracts are not suitable for multi-party contracts, e.g., the way of achieving consensus when only two negotiators are haggling over some issue is quite different if there are several negotiators involved. In the first case, a simple bargain would suffice, but in the latter a ballot process is needed. This paper presents a negotiation protocol for electronic multi-party contracts which seamlessly combines several negotiation styles. It also elaborates on the main negotiation patterns the protocol allows for: bargain (for peer-to-peer negotiation), auction (when there is competition among the negotiators) and ballot (when the negotiation aims at consensus). Finally, it describes an implementation of this protocol based on Web services, and built on the YAWL Workflow Management System.
Using choreography to support collaboration in agricultural supply chains
This paper presents an approach to support choreography in agricultural supply chains. It depicts a model for this kind of chain that considers both static and dynamic aspects, and their mapping to an underlying architecture. In particular, the model emphasizes mutual agreements, coordination of activities, quality enforcement and activity documentation. The architecture is centered on mapping chain elements to Web Services and their dynamics to the choreography of services. A case study, for soy supply chains, is used to motivate the approach.
Evaluation of a Tablet PC image annotation and retrieval tool in the parasitology domain.
The project Deployment and Assessment of an Image Annotation and Retrieval Tool has the objective of specifying and implementing an application for image support annotation and search (based on a textual and a visual description) in the biodiversity domain. This technical report presents the activities related to the use of the tablet PC tool in the parasitology domain at Unicamp. The objective of this tool is to help the comparison of morphological characteristics among different species. The report is divided into activities accomplished, application setup and specific features, followed by experimental results and conclusion. Preliminary results showed that students regarded the tool as being very useful, contributing as an alternative learning approach.
Design and Evaluation of Techniques to Utilize Implicit rating Data in Complex Information Systems
Research in personalization, including recommender systems, focuses on applications such as in online shopping malls and simple information systems. These systems consider user profile and item information obtained from data explicitly entered by users - where it is possible to classify items involved and to make personalization based on a direct mapping from user or user group to item or item group. However, in complex, dynamic, and professional information systems, such as Digital Libraries, additional capabilities are needed to achieve personalization to support their distinctive features: large numbers of digital objects, dynamic updates, sparse rating data, biased rating data on specific items, and challenges in getting explicit rating data from users. In this report, we present techniques for collecting, storing, processing, and utilizing implicit rating data of Digital Libraries for analysis and decision support. We present our pilot study to find virtual user groups using implicit rating data. We demonstrate the effectiveness of implicit rating data for characterizing users and finding virtual user communities, through statistical hypothesis testing. Further, we describe a visual data mining tool named VUDM (Visual User model Data Mining tool) that utilizes implicit rating data. We provide the results of formative evaluation of VUDM and discuss the problems raised and plans for further studies.
A New Hybrid Clustering Approach for Image Retrieval
In this paper, we present a new Hybrid Hierarchical Clustering approach for Image Retrieval. Our method combines features from both divisive and agglomerative clustering paradigms in order to yield good-quality clustering solutions with reduced computational cost. We provide several experiments showing that our technique reduces the number of required comparisons to perform a retrieval without significant loss in effectiveness when compared to flat-based solutions.
Image Retrieval based on Color and Scale Representative Image Regions (CSIR)
Content-based image retrieval (CBIR) is a challenging task. Common techniques use only low-level features. However, these solutions can lead to the so-called ‘semantic gap’ problem: images with high feature similarities may be different in terms of user perception. In this paper, our objective is to retrieve images based on color cues which may present some affine transformations. For that, we present CSIR: a new method for comparing images based on discrete distributions of distinctive color and scale image regions. We validate the technique using images with a large range of viewpoints, partial occlusion, changes in illumination, and various domains.
Wavelet-based Feature Extraction for Fingerprint Image Retrieval
This paper presents a novel approach to fingerprint retrieval for personal identification by joining three image retrieval tasks, namely, feature extraction, similarity measurement, and feature indexing, into a wavelet-based fingerprint retrieval system. We propose the use of different types of Wavelets for representing and describing the textural information present in fingerprint images. For that purposes, the feature vectors used to characterize the fingerprints are obtained by computing the mean and the standard deviation of the decomposed images in the Wavelet domain. These feature vectors are used to retrieve the most similar fingerprints given a query image, while their indexation is used to reduce the search spaces of image candidates. The different types of Wavelets used in our study include: Gabor Wavelets (GWs), Tree-Structured Wavelet Decomposition using both Orthogonal Filter Banks (TOWT) and Bi-orthogonal Filter Banks (TBOWT), as well as the Steerable Wavelets. To evaluate the retrieval accuracy of the proposed approach, a total number of eight different data sets were used. Experiments also evaluated different combinations of Wavelets with six similarity measures. The results show that the Gabor Wavelets combined with the Square Chord similarity measure achieves the best retrieval effectiveness.
A genetic programming approach for relevance feedback in region-based image retrieval systems.
This paper presents a new relevance feedback method for content-based image retrieval using local image features. This method adopts a genetic programming approach to learn user preferences and combine the region similarity values in a query session. Experiments demonstrate that the proposed method yields more effective results than the Local Aggregation Pattern (LAP)-based relevance feedback technique.
An Annotation Propagation Mechanism for Multimedia Content
Scientific research is producing and consuming large volumes of multimedia data at an ever growing rate. Metadata -- data about data -- is the primary mechanism through which context is associated to content to enhance content management. It also makes it easier to interpret and share data and helps digital curation. However, raw data often needs to go through complex processing steps before it can be consumed. During these transformation processes, original metadata from the production phase is often discarded or ignored, since its usefulness is usually limited to the first transformation step. New metadata must be associated with the final product, a time consuming task often carried out manually. Systematically associating new metadata to the result of each data transformation step is know as metadata evolution or annotation propagation. This paper introduces techniques for semantically enhancing metadata and automatically transforming them along with the data transformation processes. This helps the construction of new annotated multimedia data sets, preserving contextual information. The solution is based on: (i) the notion of semantic annotations, which are metadata structures enriched with domain ontologies; (ii) a set of transformations rules, based on ontological relations; and, (iii) workflows, which steer the sequence of transformations.
Report on WIT08 - II Workshop on Women in Information Technology
WIT is an initiative of the Brazilian Computer Society (SBC) to discuss gender issues in Information Technology (IT) in Brazil – success stories, policies to foster participation, and ways and means to attract and involve the young, especially women, in IT-related careers. Organized around guest speakers and panels, the workshop concentrated on debating problems related with women´s access to IT – the job market, digital inclusion and literacy. WIT was organized by Claudia Bauzer Medeiros and Karin Breitman, both CS faculty and part of the board of SBC. This report was submitted to the ACM-W CIS newsletter of August 2008
A framework based on semantic Web services and AI planning for the management of bioinformatics scientific workflows
Bioinformatics activities are growing all over the world, with proliferation of data and tools. This brings new challenges, such as how to understand and organize these resources, how to exchange and reuse successful experimental procedures, tools and data, and how to provide interoperability among data and tools across different sites, and for distinct user profiles. This paper describes an effort towards these directions. It is based on combining research on databases, AI and scientific workflows, on the Semantic Web, to design, reuse, annotate and document bioinformatics experiments or parts thereof. The resulting framework allows the integration of heterogeneous data and tools, and the design of experiments as scientific workflows, which are stored in databases. Moreover, it takes advantage of the notion of planning in AI to support automatic or interactive composition of tasks. These ideas are being implemented in a prototype and validated on real bioinformatics data.
Providing homogeneous access for sensor data management
We are facing the proliferation of several kinds of sensing devices, from satellites to tiny sensors. This has opened up new possibilities for us to understand, manage and monitor a given environment, from the small -- e.g., a room -- to the large -- e.g., the planet. This, however, has added a new dimension to the classic problem of heterogeneous data management -- how to handle increasing volumes of sensing data from a wide range of sensors. This report is concerned with the problem of sensor data publication. Our solution involves the design and implementation of a framework for sensor data management, which applies technologies based on Semantic Web standards, components and scientific workflows. Individual sensors or networks are encapsulated into a specific kind of component -- DCC -- which supports homogeneous access to data and software. DCCs are themselves handled by scientific workflows that provide facilities for controlling data production, integration and publication. As a result, applications that require sensor data will instead interact with workflows, being liberated from concerns such as sensor particularities, or provide separate handlers for real time streams. The report also presents initial implementation results.
Integrating image and spatial data for biodiversity information management
Biologists gather many kinds of data for biodiversity studies; these data are managed by distinct types of information systems. GIS-based biodiversity systems support sophisticated spatial correlations on living beings and their habitats, and spatio-temporal ecosystem modeling. Image information systems allow content-based image retrieval, to help species identification based on similarity (e.g., shape and color characteristics). Different kinds of rule-based systems support species characterization. Unfortunately, these systems (and the underlying data) are independent of each other. This paper presents a solution that seamlessly combines these functionalities, supporting queries that merge textual descriptions, spatial correlations and content-based predicates. The solution is being implemented at Virginia Tech, for identification and data retrieval, supporting management of fish species. It takes advantage of innovations in Digital Library technology to combine networked collections of heterogeneous data under integrated management.
Contour Salience Descriptors for Effective Image Retrieval and Analysis.
This work exploits the resemblance between content-based image retrieval and image analysis with respect to the design of image descriptors and their effectiveness. In this context, two shape descriptors are proposed: contour saliences and segment saliences. Contour saliences revisits its original definition, where the location of concave points was a problem, and provides a robust approach to incorporate concave saliences. Segment saliences introduces salience values for contour segments, making it possible to use an optimal matching algorithm as distance function. The proposed descriptors are compared with convex contour saliences, curvature scale space, and beam angle statistics using a fish database with 11,000 images organized in 1,100 distinct classes. The results indicate segment saliences as the most effective descriptor for this particular application and confirm the improvement of the contour salience descriptor in comparison with convex contour saliences.
Incorporating Time in an Object Oriented Database System (in portuguese)
Este trabalho descreve a Camada de Gerenciamento Temporal, um subsistema de gerenciamento temporal implementado para o banco de dados orientado a objetos O2. Este subsistema permite a definição de esquemas temporais e a manipulação (consultas e atualizações) de objetos nesses esquemas. Estes comandos são traduzidos pelo subsistema em programas e consultas executados pelo banco de dados subjacente. O trabalho apresentado contribui para a discussão sobre a implementação de sistemas temporais orientados a objetos, pouco explorado na literatura.
Using Versions in GIS
Geographic information systems GIS have become important tools in public planning activities (e.g, in environmental or utility management). This type of activity requires the creation and management of alternative scenarios, as well as analysis of temporal data evolution. Existing systems provide limited support for these operations, and appropriate tools are yet to be developed. This paper presents a solution to this problem. This solution is based on managing temporal data and alternatives using the DBV version mechanism. It provides efficient handling and storage of versions, and supports the creation of alternatives for decision-making activities. A reduced version of this report appeared in the Proceedings of the DEXA'94 Conference --- 5th International Conference on Database and Expert Systems Applications, Athens, Greece.
Methodologies for Schema Conversion in Heterogeneous Database Systems (in portuguese)
Sistemas de Bancos de Dados Heterogêneos (SBDHs) são sistemas que integram, num ambiente cooperativo, sistemas de bancos de dados (SBDs) autônomos e heterogêneos entre si, com relação à semântica dos seus dados e/ou às características dos seus SGBDs (modelo de dados, linguagens de manipulação de dados e aspectos de implementação). Uma propriedade desejável nesses sistemas é a transparência de modelos, que permite ao usuário enxergar e manipular os dados localizados em diferentes SBDs, através do modelo e da linguagem de manipulação de dados que ele utilizava no seu SBD local, antes do mesmo ser incorporado ao SBDH. A propriedade em questão é conseguida através do mapeamento entre os dados e operações dos SBDs componentes. Este trabalho apresenta metodologias para conversão de esquemas em SBDHs construídos para integrar SBDs que utilizam o modelo de dados rede ou o modelo de dados relacional. Essas metodologias são importantes para obtenção da transparência de modelos.
Rule Application in GIS - a Case Study
Production rules in database systems have been used mostly for integrity-related issues (e.g., derived data maintenance, authority checking and constraint verification). This paper analyzes the need for using production rules in geographic information systems, for a special family of applications---utility management systems. This framework is applied to a real life large scale application---the development of an integrated database system for the maintenance and expansion of the telephone network in Brazil.
Managing Time in Object-Oriented Databases
This paper presents a new approach for modelling and querying temporal object oriented databases. The model presented in this paper extends previous work in the area, by supporting the evolution of all object properties through time (inheritance, composition and behavior), and allowing temporal schema evolution. A prototype of this model is being implemented as a time-managing layer on top of the O2 object-oriented database system. In order to manipulate our temporal objects, we have extended the O2 query language with temporal constructs, which we also discuss in the paper.
Modelling Geographic Information Systems using an Object Oriented Framework
Geographic information systems demand the processing of complex data using specialized operations, not available in traditional database systems. Even though there exist commercial systems that provide some of these facilities, there is a lack of proper support, which should cover not only the implementation but also the design stage. This paper answers this latter need, discussing the steps for modelling databases for geographic information systems using the paradigm of object orientation.
Browsing and Querying Object-Oriented Databases
We present a new interface for Object-Oriented Database Management Systems (OODBMSs). The GOODIES system (an acronym for Graphical Object Oriented Database Interface with Extended Synchronism) combines and expands the functions of many existing interface systems, introducing some new concepts for improved browsing in an OODBMS. The implementation of GOODIES proposes a new approach to database interfaces development: instead of being strongly dependent of the underlying DBMS, GOODIES is based on the main features of the object-oriented data model. The system design is based on an internal model and on an external model. The internal model ,defines the relationships that bind the interface to the DBMS. The external model determines the possible interaction between the user and the interface system. This paper describes the concepts of the external model of the GOODIES system.
Maintaining Integrity Constraints across Versions in a Database
This paper analyzes the problem of maintaining application-dependent integrity constraints in databases for design enviroments, Such enviroments are characterized by the need to support different types of interaction between integreity maintenance and version maintenance nechanisms. The paper describes these problems, and propose a framework in which they can be treated homogeneously. We thus bridge the gap existing between research on constraint maintenance and on version control, which has so far posed several problems to researchers in these two areas.
Implementing Integrity Control in Active Databases
This paper presents an integrity maintenance system that has been developed for maintainning static constrais in databases, using the active database paradigm. This system has beem added to O2 object oriented database system, and is fully funcional. Constraints are specified by the user in a first order logic language, and transformed in production rules, which are stored in the database. The rules are then used to maintain the corresponding set of constraints, for all applications that use the database, and which no longer need to worry about integrity control. We extend previous work on constraint maintenance in two ways: our system can be used as a constraint maintenance layer on top of object-oriented, relational and nested relational databases; in the case of object-oriented systems, we provide constraints support not only in the case of object composition, but also consider inheritance and methods.
Scientific Workflow Management: WASA Architecture and Applications
 
Image retrieval with relevance feedback based on genetic programming
In the last years, large digital image collections are generated, manipulated, and stored in databases. In this scenery, it is very important to develop mechanisms to provide automatic means to retrieve images in an efficient and effective way. However, the subjectivity of the user perception of an image usually hampers a fully automatic search and retrieval. Relevance Feedback is one of the commonest approaches to overcome this difficult. In this paper, a new content-based image retrieval framework with relevance feedback is proposed. This framework uses Genetic Programming (GP) to learn the user needs. The objective of this learning method is to find a function that combines different values of similarity, from distinct descriptors, and best encodes the user perception of image similarity. Several experiments are performed to validate the proposed method, aiming to compare our work with other relevance feedback techniques. The experiment results show that the proposed method outperforms all of them.
Integrating Heuristics and Spatial Databases - a case study
This paper presents part of the ongoing efforts at ic­unicamp to apply heuristic algorithms to vectorial georeferenced data in order to help decision support in urban planning. The results reported are original in the sense that they combine recent re­ search in both combinatorial algorithm development and geographic databases, using them in the solution of a practical problem. A first prototype, described in the paper, has already been developed and tested against real data on the city of Campinas, to support planning activities for the São Paulo State Post Office System, Brazil.
User Interface Issues in Geographic Information Systems
Recently, much research effort has been employed in the area of Geographic Infor­ mation Systems due to the vast potential for applications of this technology. Simul­ taneously, user interface subsystems of software products have received attention since the interface has marked influence in software acceptance. This paper presents an over­ view of research done in the intersection of these areas. The main approaches and the current problems of user interfaces for Geographical Information Systems are discussed and analyzed. This study concludes with open problems and new research directions for future work in this area.
Designing and Implementing the User Interface of Geographic Digital Libraries
Geographic data are useful for a large set of applications, such as urban planning and environmental control. These data are, however, very expensive to acquire and maintain. Moreover, their use is often restricted, for lack of dissemination mechanisms. Digital libraries are a good approach for increasing data availability and therefore re­ ducing cost, since they provide efficient storage and access to large volumes of data. Geographic applications can diminish their costs by reusing and sharing data through Geographic Digital Libraries (gdl). One major drawback to this approach is that it creates the necessity of providing facilities for a large and heterogeneous community of users to search and interact with these Geographic Libraries. We present a solu­ tion for this problem, based on a framework that allows the design and construction of customizable user interfaces for gdl applications. This framework relies on two main concepts: a Geographic User Interface Architecture and a Geographic Digital Library
Access Structure for Moving Points
Several applications require management of data which is spatially dynamic, e.g., tracking of battle ships or moving cells in a blood sample. The capability of handling the temporal aspect, i.e., the history of such type of data, is also important. This paper presents and evaluates three temporal extensions of the R-tree, the 3D R-tree, the 2+3 R-tree and the HR-tree, which are capable of indexing spatiotemporal data. Our experiments have shown that the while the HR-tree was the larger structure, its query processing cost was over 50% smaller than the ones yielded by the 3D R-tree and the 2+3 R-tree. Also compared to the (non-practical) approach of storing one R-tree for each of the spatial database states it offered the same query processing cost, saving around one third of storage space.
Workflow Management in Geoprocessing Scientific Applications
This paper presents a system which is being developed in the University of Münster to support scientific application environments. This system -- WASA -- is based on taking advantage of workflows to document and monitor the execu­ tion of scientific applications. A geoprocessing application is used throughout the paper to illustrate and justify the specificity of the problem and our proposed solution.
An Extensible Framework for Temporal Database Applications
There is a wide range of scientific application domains requiring sophisticated management of spatio-temporal data. However, existing database management systems offer very limited (if any at all) support for managing such data. Thus, it is left to the researchers themselves to repeatedly code this management into each application. Besides being a time consuming task, this process is bound to introduce errors and increase the complexity of application management and data evolution. This paper addresses this very point. We present an extensible framework, based on extending an object-oriented database system, with kernel spatio-temporal classes, data structures and functions, to provide support for the development of spatio-temporal applications. Even though the paper’s arguments are centered on geographic applications, the proposed framework can be used in other application domains where spatial and temporal data evolution must be considered (e.g., Biology).
A Graph-based Approach for Multiscale Shape Analysis.
This paper presents the advantages of computing two recently proposed shape descriptors, multiscale fractal dimension and contour saliences, using the image foresting transform---a graph-based approach to the design of image processing operators. It introduces a robust approach to estimate contour saliences (peaks of high curvature) by exploiting the relation between contour and skeleton. The paper also compares both shape descriptors to fractal dimension, Fourier descriptors, and moment invariants with respect to their invariance to object characteristics that belong to a same class (compact-ability) and to their discriminatory ability to separate objects that belong to distinct classes (separability).
An OAI-based Digital Library Framework for Biodiversity Information Systems
Biodiversity information systems (BISs) involve all kinds of heterogeneous data, which include ecological and geographical features. However, available information systems offer very limited support for managing such data in an integrated fashion, and integration is often based on geographic coordinates alone. Furthermore, such systems do not fully support image content management (e.g., photos of landscapes or living organisms), a requirement of many BIS end-users. In order to meet their needs, these users - e.g., biologists, environmental experts - often have to alternate between distinct biodiversity and image information systems to combine information extracted from them. This cumbersome operational procedure is forced on users by lack of interoperability among these systems. This hampers the addition of new data sources, as well as cooperation among scientists. The approach provided in this paper to meet these issues is based on taking advantage of advances in Digital Library (DL) innovations to integrate networked collections of heterogeneous data. It focuses on creating the basis for a biodiversity information system under the digital library perspective, combining new techniques of content-based image retrieval and database query processing mechanisms. This approach solves the problem of system switching, and provides users with a flexible platform from which to tailor a BIS to their needs.
A Direct Manipulation User Interface for Querying Geographic Databases
This paper presents an architecture for a direct manipulation user interface for browsing and querying geographic data. This interface rovides users with a high level object oriented conceptual view of the underlying database, independent of the database's native data model. It lets users manipulate different representations of a single georeferenced entity, thereby adding a new degree of flexibility to querying facilities.
WASA - a Workflow-based Architecture to Support Scientific Database Applications
Scientific applications which involve the use of databases are of emerging in­terest. However, computer­based environments which support such applications are still in their infancy. The overall goal of the WASA project is to provide such an environment, and to do so with the specific domains of geosciences and bio­computing in mind. This paper introduces the architecture of WASA, explains its design decisions, describes how its various components interact, and how it can be utilized in the target applications. One important design decision has been to re­use existing database technology as far as possible, but to integrate this with the emerging paradigm of workflow management. As a result, WASA has a workflow manager as its central component; this tool manages information as well as other tools associated with a given environment, some of which are domain­independent, while others have to be adapted whenever the system is used in a new application area.
A New Framework to Combine Descriptors for Content-Based Image Retrieval
Methods that combine image database descriptors have strong influence on the effectiveness of content-based image retrieval (CBIR) systems. Although there are many combination functions described in the image processing literature, empirical evaluation studies have shown that those functions do not perform consistently well across different contexts (queries, image collections, users). Moreover, it is often very difficult for human beings to identify optimal combination functions for a particular application. In this paper, we propose a novel framework using {\em Genetic Programming} to combine image database descriptors for CBIR. Our framework is validated through several experiments involving two image databases and a specific domain, where the images are retrieved based on the shape of their objects.
A Survey on Information Systems Interoperability
The interoperability of information systems has
 

Powered by Plone CMS, the Open Source Content Management System

This site conforms to the following standards: