Matching entries: 0
settings...
Author Title Year Journal/Proceedings Reftype DOI/URL
Hernández, J.F., dos Santos, J.A. and da S. Torres, R. Learning to Combine Spectral Indices with Genetic Programming 2016 29th SIBGRAPI -- Conference on Graphics, Patterns and Images  inproceedings  
Abstract: This paper introduces a Genetic Programming-based method for band selection and combination, aiming to support remote sensing image classification tasks. Relying on ground-truth data, our method selects spectral bands and finds the arithmetic combination of those bands (i.e., spectral index) that best separates examples of different classes. Experimental results demonstrate that the proposed method is very effective in pixel-wise binary classification problems.

BibTeX:
@inproceedings{Hernandez2016SIBGRAPI,
  author = {Juan Felipe Hernández and Jefersson Alex dos Santos and Ricardo da S. Torres},
  title = {Learning to Combine Spectral Indices with Genetic Programming},
  booktitle = {29th SIBGRAPI -- Conference on Graphics, Patterns and Images},
  year = {2016}
}
      


Mariano, G., Staggemeier, V.G., Morellato, L.P., da S. Torres, R. and Alberton, B. A Visual Rhythm Approach to Visualize Multidimensional Cyclical Data: A Case Study on Phenological Data Analysis 2016 International Conference on Digital Libraries  inproceedings  
Abstract: Phenology is an important science that investigates the recurrent life cycles from living beings
(plants and animals) and their relations to environmental changes. Traditional phenology studies
typically are based on observations made by experts in the field over time and usually are
correlated with climate data collected by weather sensors. In more recent studies, phenology
experts have also been using images to monitor vegetation remotely. Although several
approaches have been proposed for visualizing time series associated with phenology data, they
are not suitable due to the lack of mechanisms: (i) to handle long-term series with many variables
of different data types, and (ii) to identify cyclical temporal patterns using both images and
numerical data. In this paper, we present a novel approach to visualize phenological data by
combining radial visual structures along with visual rhythms. Radial visual structures are used to
provide contextual insights regarding cyclical phenomena, while the visual rhythm encoding is
used to summarize long-term time series into compact representations. Our objective is to
support the discovery process by allowing the analysis of multidimensional cyclical data. This
paper also discusses the use of the proposed visual structures in a case study concerning the
analysis of data produced in the context of the e-phenology Project.

BibTeX:
@inproceedings{Mariano2016ICDL,
  author = {Greice Mariano and Vanessa G. Staggemeier and Leonor Patrícia Morellato and Ricardo da S. Torres and Bruna Alberton},
  title = {A Visual Rhythm Approach to Visualize Multidimensional Cyclical Data: A Case Study on Phenological Data Analysis},
  booktitle = {International Conference on Digital Libraries},
  year = {2016}
}
      


Nogueira, K., dos Santos, J.A., Fornazari, T., Silva, T.S.F., Morellato, L.P.C. and da S. Torres, R. Towards vegetation species discrimination by using data-driven descriptors 2016 9th IAPR Workshop on Pattern Recognition in Remote Sensing  conference  
Abstract: In this paper, we analyse the use of Convolutional Neural Networks (CNNs or ConvNets) to discriminate vegetation species with few labelled samples.
To the best of our knowledge, this is the first work dedicated to the investigation of the use of deep features in such task.
The experimental evaluation demonstrate that deep features significantly outperform well-known feature extraction techniques.
The achieved results also show that it is possible to learn and classify vegetation patterns even with few samples.
This makes the use of our approach feasible for real-world mapping applications, where it is often difficult to obtain large training sets.

BibTeX:
@conference{Nogueira2016PRRS,
  author = {Keiller Nogueira and Jefersson Alex dos Santos and Tamires Fornazari and Thiago S. F. Silva and Leonor Patricia C. Morellato and Ricardo da S. Torres},
  title = {Towards vegetation species discrimination by using data-driven descriptors},
  booktitle = {9th IAPR Workshop on Pattern Recognition in Remote Sensing},
  year = {2016}
}
      


Muñoz, J.A.V., da Silva Torres, R. and Gonçalves, M.A. A Soft Computing Approach for Learning to Aggregate Rankings 2015 Proceedings of the 24th ACM International on Conference on Information and Knowledge Management, CIKM, pp. 83-92  inproceedings DOI URL 
Abstract: This paper presents an approach to combine rank aggregation techniques using a soft computing technique -- Genetic Programming -- in order to improve the results in Information Retrieval tasks. Previous work shows that by combining rank aggregation techniques in an agglomerative way, it is possible to get better results than with individual methods. However, these works either combine only a small set of lists or are performed in a completely ad-hoc way. Therefore, given a set of ranked lists and a set of rank aggregation techniques, we propose to use a supervised genetic programming approach to search combinations of them that maximize effectiveness in large search spaces. Experimental results conducted using four datasets with different properties show that our proposed approach reaches top performance in most datasets. Moreover, this cross-dataset performance is not matched by any other baseline among the many we experiment with, some being the state-of-the-art in learning-to-rank and in the supervised rank aggregation tasks. We also show that our proposed framework is very efficient, flexible, and scalable.

BibTeX:
@inproceedings{Munoz2015CIKM,
  author = {Javier Alvaro Vargas Muñoz and Ricardo da Silva Torres and Marcos André Gonçalves},
  title = {A Soft Computing Approach for Learning to Aggregate Rankings},
  booktitle = {Proceedings of the 24th ACM International on Conference on Information and Knowledge Management, CIKM},
  year = {2015},
  pages = {83--92},
  url = {http://doi.acm.org/10.1145/2806416.2806478},
  doi = {http://doi.org/10.1145/2806416.2806478}
}
      


Nicastro, F., Pereira, R., Alberton, B., Morellato, L.P.C., Baranauskas, C. and da Silva Torres, R. A Semiotic-informed Approach to Interface Guidelines for Mobile Applications - A Case Study on Phenology Data Acquisition 2015
Vol. Volume 3Proceedings of the 17th International Conference on Enterprise Information Systems (ICEIS), pp. 34-43 
inproceedings DOI URL 
Abstract: Portable devices have been experimented for data acquisition in different domains, e.g., logistics and census data acquisition. Nevertheless, their large-scale adoption depends on the development of effective applications with a careful interaction design. In this paper, we revisit existing interface design strategies and propose a guideline composed of semiotic-informed rules and questions for mobile user interface design. We demonstrate the use of the guideline in the evaluation of mobile application interfaces proposed for phenological data acquisition in the field.

BibTeX:
@inproceedings{Nicastro2015ICEIS,
  author = {Flavio Nicastro and Roberto Pereira and Bruna Alberton and Leonor Patricia C. Morellato and Cecilia Baranauskas and Ricardo da Silva Torres},
  title = {A Semiotic-informed Approach to Interface Guidelines for Mobile Applications - A Case Study on Phenology Data Acquisition},
  booktitle = {Proceedings of the 17th International Conference on Enterprise Information Systems (ICEIS)},
  year = {2015},
  volume = {Volume 3},
  pages = {34--43},
  url = {http://dx.doi.org/10.5220/0005379600340043},
  doi = {http://doi.org/10.5220/0005379600340043}
}
      


Okada, C.Y., Pedronette, D.C.G. and da Silva Torres, R. Unsupervised Distance Learning by Rank Correlation Measures for Image Retrieval 2015 Proceedings of the 5th ACM on International Conference on Multimedia Retrieval, pp. 331-338  inproceedings DOI URL 
Abstract: Ranking accurately collection images is the main objective of Content-based Image Retrieval (CBIR) systems. In fact, the set of images ranked at the first positions generally defines the effectiveness of provided search services, i.e., they are used for assessing automatically the quality of search systems as this set usually contains the collection images that are of interest. Recently, the use of ranking information (e.g., rank correlation) has been used in different research initiatives with the objective of improving the effectiveness of image retrieval tasks. This paper presents a broad rank correlation analysis for unsupervised distance learning on image retrieval tasks. Various well-known rank correlation measures are considered and two new measures are proposed. Several experiments were conducted considering various image datasets involving shape, color, and texture descriptors. Experimental results demonstrate that ranking information can be exploited for distance learning tasks successfully. Evaluated approaches yield better results in terms of effectiveness than various state-of-the-art algorithms.

BibTeX:
@inproceedings{Okada2015ICMR,
  author = {César Yugo Okada and Daniel C. G. Pedronette and Ricardo da Silva Torres},
  title = {Unsupervised Distance Learning by Rank Correlation Measures for Image Retrieval},
  booktitle = {Proceedings of the 5th ACM on International Conference on Multimedia Retrieval},
  publisher = {ACM},
  year = {2015},
  pages = {331--338},
  url = {http://doi.acm.org/10.1145/2671188.2749335},
  doi = {http://doi.org/10.1145/2671188.2749335}
}
      


Pedronette, D.C.G. and da Silva Torres, R. Unsupervised Effectiveness Estimation for Image Retrieval Using Reciprocal Rank Information 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, pp. 321-328  inproceedings DOI URL 
Abstract: In this paper, we present an unsupervised approach for estimating the effectiveness of image retrieval results obtained for a given query. The proposed approach does not require any training procedure and the computational efforts needed are very low, since only the top-k results are analyzed. In addition, we also discuss the use of the unsupervised measures in two novel rank aggregation methods, which assign weights to ranked lists according to their effectiveness estimation. An experimental evaluation was conducted considering different datasets and various image descriptors. Experimental results demonstrate the capacity of the proposed measures in correctly estimating the effectiveness of different queries in an unsupervised manner. The linear correlation between the proposed and widely used effectiveness evaluation measures achieves scores up to 0.86 for some descriptors.

BibTeX:
@inproceedings{Pedronette2015SIBGRAPI,
  author = {Daniel C. G. Pedronette and Ricardo da Silva Torres},
  title = {Unsupervised Effectiveness Estimation for Image Retrieval Using Reciprocal Rank Information},
  booktitle = {28th SIBGRAPI Conference on Graphics, Patterns and Images},
  year = {2015},
  pages = {321--328},
  url = {http://dx.doi.org/10.1109/SIBGRAPI.2015.28},
  doi = {http://doi.org/10.1109/SIBGRAPI.2015.28}
}
      


Pinto-Cáceres, S.M., Almeida, J., Baranauskas, M.C.C. and da Silva Torres, R. FISIR: A Flexible Framework for Interactive Search in Image Retrieval Systems 2015
Vol. Part {I}MultiMedia Modeling - 21st International Conference, MMM, pp. 335-347 
inproceedings DOI URL 
Abstract: This paper presents a flexible framework for interactive search in
image retrieval systems. Our approach allows for the visual structure
change in order to produce coherent layouts, which highlight the
most relevant results according to user needs. This innovative framework
is flexible in the sense it supports the dynamic creation of several
hybrid visual designs, based on the combination of different visualization
strategies. Results from a subjective evaluation demonstrate that
the dynamic hybrid layouts created by the proposed framework provide
the end-users with an effective user interface for intuitive browsing
and searching experience.

BibTeX:
@inproceedings{Pinto-Caceres2015MMM,
  author = {Sheila M. Pinto-Cáceres and Jurandy Almeida and Maria Cec\ilia Calani Baranauskas and Ricardo da Silva Torres},
  title = {FISIR: A Flexible Framework for Interactive Search in Image Retrieval Systems},
  booktitle = {MultiMedia Modeling - 21st International Conference, MMM},
  year = {2015},
  volume = {Part I},
  pages = {335--347},
  url = {http://dx.doi.org/10.1007/978-3-319-14445-0_29},
  doi = {http://doi.org/10.1007/978-3-319-14445-0_29}
}
      


Pisani, F., Pedronette, D.C.G., da Silva Torres, R. and Borin, E. Improving the Performance of the Contextual Spaces Re-Ranking Algorithm on Heterogeneous Systems 2015 XVI Simpósio em Sistemas Computacionais de Alto Desempenho (WSCAD)  inproceedings  
Abstract: Re-ranking algorithms have been proposed to improve the effectiveness of Content-Based Image Retrieval (CBIR) systems by exploiting contextual information encoded in distance measures and ranked lists. In this paper, we show how we improved the efficiency of one of these algorithms, called Contextual Spaces Re-Ranking. We propose a modification to the algorithm that reduces its execution time by 1.6 x on average and improves its accuracy in most of our test cases. We also parallelized the implementation with OpenCL to use the CPU and GPU of an Accelerated Processing Unit (APU). Employing these devices to run different parts of the code resulted in speedups that range from
3.3 x to 4.2 x in comparison with the total execution time of the serial version.

BibTeX:
@inproceedings{Pisani2015WSCAD,
  author = {Flavia Pisani and Daniel C. G. Pedronette and Ricardo da Silva Torres and Edson Borin},
  title = {Improving the Performance of the Contextual Spaces Re-Ranking Algorithm on Heterogeneous Systems},
  booktitle = {XVI Simpósio em Sistemas Computacionais de Alto Desempenho (WSCAD)},
  year = {2015}
}
      


Valem, L.P., Pedronette, D.C.G., da Silva Torres, R., Borin, E. and Almeida, J. Effective, Efficient, and Scalable Unsupervised Distance Learning in Image Retrieval Tasks 2015 Proceedings of the 5th ACM on International Conference on Multimedia Retrieval, pp. 51-58  inproceedings DOI URL 
Abstract: Various unsupervised learning methods have been proposed with significant improvements in the effectiveness of image search systems. However, despite the relevant effectiveness gains, these approaches commonly require high computation efforts, not addressing properly efficiency and scalability requirements. In this paper, we present a novel unsupervised learning approach for improving the effectiveness of image retrieval tasks. The proposed method is also scalable and efficient as it exploits parallel and heterogeneous computing on CPU and GPU devices. Extensive experiments were conducted considering ve different public image collections and several descriptors. This rigorous experimental protocol evaluates the effectiveness, efficiency, and scalability of the proposed approach, and compares it with previous methods. Experimental results demonstrate that high effectiveness gains (up to +29 can be obtained requiring small run times.

BibTeX:
@inproceedings{Valem2015ICMR,
  author = {Lucas Pascotti Valem and Daniel C. G. Pedronette and Ricardo da Silva Torres and Edson Borin and Jurandy Almeida},
  title = {Effective, Efficient, and Scalable Unsupervised Distance Learning in Image Retrieval Tasks},
  booktitle = {Proceedings of the 5th ACM on International Conference on Multimedia Retrieval},
  year = {2015},
  pages = {51--58},
  url = {http://doi.acm.org/10.1145/2671188.2749336},
  doi = {http://doi.org/10.1145/2671188.2749336}
}
      


Waku, G.M., Bollis, E.R., Rubira, C.M.F. and da Silva Torres, R. A Robust Software Product Line Architecture for Data Collection in Android Platform 2015 IX Brazilian Symposium on Components, Architectures and Reuse Software (SBCARS), pp. 31-39  inproceedings DOI  
Abstract: Android is an open platform, developed by Google and Open Mobile Handset Alliance targeting mobile devices. Its constant evolution and increasing cost reduction made them suitable for complex applications especially for data collection applications. Data collection is a domain which evolved to use mobile devices to collect information, targeting different fields of study including: physical and social sciences, humanities, business, demographic surveys, agriculture, biology, and geology. Android usually runs on different hardware and software domains with similar functional and non-functional features. The data collection domain has a lot of sub-domains, creating an opportunity to explore software variability and quality properties such as reliability, availability, and data integrity using Component-Based Development (CBD), Fault Tolerance Techniques and Software Product Line (SPL) with Aspect-Oriented Software Development (AOSD). However, the use of mobile application for data collection poses some challenges like severe hardware restrictions (such as limited power processing and short battery lifetime) and the use of sophisticated techniques can negatively impact in application performance, and quality properties. In this work, these issues were addressed by proposing the development of a robust SPL architecture called Robust SPL for Data Collection (R-SPL-DC) and a real application called E-Phenology Collector for data collection domain to assess the use fault tolerance techniques, CBD, SPL, and AOSD to ensure availability, reliability, and data integrity without significant impacts on the overall performance of the mobile device. The results have shown that the use of R-SPL-DC is promising and suits the requirements for data collection domain.

BibTeX:
@inproceedings{Waku2015SBCARS,
  author = {Gustavo M. Waku and Edson R. Bollis and Cecilia M. F. Rubira and Ricardo da Silva Torres},
  title = {A Robust Software Product Line Architecture for Data Collection in Android Platform},
  booktitle = {IX Brazilian Symposium on Components, Architectures and Reuse Software (SBCARS)},
  year = {2015},
  pages = {31-39},
  doi = {http://doi.org/10.1109/SBCARS.2015.14}
}
      


Calumby, R.T., da Silva Torres, R. and calves , M.A.G. Diversity-driven Learning for Multimodal Image Retrieval with Relevance Feedback 2014 IEEE International Conference on Image Processing  inproceedings  
Abstract: We introduce a new genetic programming approach for enhancing the
user search experience based on relevance feedback over results produced
by a multimodal image retrieval technique with explicit diversity
promotion. We have studied maximal marginal relevance re-ranking
methods for result diversification and their impacts on the overall
retrieval effectiveness. We show that the learning process using
diverse results may improve user experience in terms of both the
number of relevant items retrieved and subtopic coverage.

BibTeX:
@inproceedings{Calumby2014ICIP,
  author = {Rodrigo Tripodi Calumby and Ricardo da Silva Torres and Marcos A. Gon¸ calves},
  title = {Diversity-driven Learning for Multimodal Image Retrieval with Relevance Feedback},
  booktitle = {IEEE International Conference on Image Processing},
  year = {2014}
}
      


Conti, J., Faria, F.A., Almeida, J., Alberton, B., Morellato, L.P., Camolesi, L. and da Silva Torres, R. Evaluation of Time Series Distance Functions in the Task of Detecting Remote Phenology Patterns 2014 22nd International Conference on Pattern Recognition, pp. 3126-3131  inproceedings  
Abstract: Phenology is the study of periodic natural phenomena and their relationship
to climate. Usually, phenology studies consider the identification
of patterns on temporal data. In those studies, several phenological
change patterns are often encoded in time series for analysis and
knowledge extraction. In this paper, we evaluate the effectiveness
of several time series similarity functions in the task of classifying
time series related to phenological phenomena characterized by near-surface
vegetation indices extracted from images. In addition, we performed
a correlation analysis to identify potential candidates for combination.

BibTeX:
@inproceedings{Conti2014ICPR,
  author = {José Conti and Fabio A. Faria and Jurandy Almeida and Bruna Alberton and Leonor Patrícia Morellato and Luis Camolesi and Ricardo da Silva Torres},
  title = {Evaluation of Time Series Distance Functions in the Task of Detecting Remote Phenology Patterns},
  booktitle = {22nd International Conference on Pattern Recognition},
  year = {2014},
  pages = {3126-3131}
}
      


Faria, F.A., Rocha, A. and da Ricardo, S. A Framework for Pattern Classifier Selection and Fusion 2014 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases  inproceedings  
Abstract: In this work, we propose a framework for classifier selection and
fusion. Our method seeks to combine image characterization and learning
methods by means of a meta-learning approach responsible for assessing
which methods contribute more towards the solution of a given problem.
The framework uses three different strategies of classifier selection
that pinpoint the less correlated, yet effective, classifiers through
a series of diversity measure analysis. The experiments show that
the proposed approaches yield comparable results to well-known algorithms
from the literature on many different applications but using less
learning and description methods as well as not incurring in the
curse of dimensionality and normalization problems common to some
fusion techniques. Furthermore, our approach yields effective classification
results using very reduced training sets.

BibTeX:
@inproceedings{Faria2014ECML-PKDD,
  author = {Fabio A. Faria and Anderson Rocha and Ricardo da S.},
  title = {A Framework for Pattern Classifier Selection and Fusion},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year = {2014}
}
      


Pedronette, D.C.G. and da S. Torres, R. Unsupervised Manifold Learning by Correlation Graph and Strongly Connected Components for Image Retrieval 2014 IEEE International Conference on Image Processing  inproceedings  
Abstract: This paper presents a novel manifold learning approach that takes
into account the intrinsic dataset geometry. The dataset structure
is modeled in terms of a Correlation Graph and analyzed using Strongly
Connected Components (SCCs). The proposed manifold learning approach
defines a more effective distance among images, used to improve the
effectiveness of image retrieval systems. Several experiments were
conducted for different image retrieval tasks involving shape, color,
and texture descriptors. The proposed approach yields better results
in terms of effectiveness than various methods recently proposed
in the literature.

BibTeX:
@inproceedings{Pedronette2014ICIP,
  author = {Daniel C. G. Pedronette and Ricardo da S. Torres},
  title = {Unsupervised Manifold Learning by Correlation Graph and Strongly Connected Components for Image Retrieval},
  booktitle = {IEEE International Conference on Image Processing},
  year = {2014}
}
      


Pedronette, D.C.G., Penatti, O.A.B., Calumby, R.T. and da Silva Torres, R. Unsupervised Distance Learning By Reciprocal kNN Distance for Image Retrieval 2014 Proceedings of International Conference on Multimedia Retrieval, pp. 345:345-345:352  inproceedings DOI URL 
Abstract: This paper presents a novel unsupervised learning approach that takes
into account the intrinsic dataset structure, which is represented
in terms of the reciprocal neighborhood references found in different
ranked lists. The proposed Reciprocal kNN Distance defines a more
effective distance between two images, and is used to improve the
effectiveness of image retrieval systems. Several experiments were
conducted for different image retrieval tasks involving shape, color,
and texture descriptors. The proposed approach is also evaluated
on multimodal retrieval tasks, considering visual and textual descriptors.
Experimental results demonstrate the effectiveness of proposed approach.
The Reciprocal kNN Distance yields better results in terms of effectiveness
than various state-of-the-art algorithms.

BibTeX:
@inproceedings{Pedronette2014ICMR,
  author = {Daniel C. G. Pedronette and Otávio A. B. Penatti and Rodrigo Tripodi Calumby and Ricardo da Silva Torres},
  title = {Unsupervised Distance Learning By Reciprocal kNN Distance for Image Retrieval},
  booktitle = {Proceedings of International Conference on Multimedia Retrieval},
  publisher = {ACM},
  year = {2014},
  pages = {345:345--345:352},
  url = {http://doi.acm.org/10.1145/2578726.2578770},
  doi = {http://doi.org/10.1145/2578726.2578770}
}
      


Pedronette, D.C.G., Calumby, R.T. and da S. Torres, R. Semi-Supervised Learning for Relevance Feedback on Image Retrieval Tasks 2014 Conference on Graphics, Patterns and Images, pp. 243-250  inproceedings  
Abstract: Relevance feedback approaches have been estab- lished as an important
tool for interactive search, enabling users to express their needs.
However, in view of the growth of multimedia collections available,
the user efforts required by these methods tend to increase as well,
demanding approaches for reducing the need of user interactions.
In this context, this paper proposes a semi-supervised learning algorithm
for relevance feedback to be used in image retrieval tasks. The proposed
semi-supervised algorithm aims at using both supervised and unsupervised
approaches simultaneously. While a supervised step is performed using
the information collected from the user feedback, an unsupervised
step exploits the intrinsic dataset structure, which is represented
in terms of ranked lists of images. Several experiments were conducted
for different image retrieval tasks involving shape, color, and texture
descriptors and different datasets. The proposed approach was also
evaluated on multimodal retrieval tasks, considering visual and textual
descriptors. Experimental results demonstrate the effectiveness of
the proposed approach.

BibTeX:
@inproceedings{Pedronette2014SIBGRAPI,
  author = {Daniel C. G. Pedronette and Rodrigo T. Calumby and Ricardo da S. Torres},
  title = {Semi-Supervised Learning for Relevance Feedback on Image Retrieval Tasks},
  booktitle = {Conference on Graphics, Patterns and Images},
  year = {2014},
  pages = {243-250}
}
      


dos Santos, L.C.B., Almeida, J., Santos, J.A., Guimarães, S.J.G., de A. Araújo, A., Alberton, B., Morellato, L.P. and da Silva Torres, R. Phenological event detection by visual rhythm dissimilarity analysis 2014 10th IEEE International eScience Conference  inproceedings  
Abstract: Plant phenology has been exploited as an important research venue
for assessing the impact of climate changes. One common approach
for monitoring vegetation relies on the use of digital cameras. The
employment of imaging techniques for phenological observation allows
the extraction and analysis of visual characteristics based on color
and texture information with the objective of determining plant life
cycle changes, such as the beginning of the leaf flushing or the
senescence period. This paper presents a novel approach for detecting
phenological changes by analyzing image temporal series. Our method
is based on the use of visual rhythm analysis and the adoption of
a dissimilarity measure to detect visual changes in the time line.
Experiments were conducted on a three-year data set composed of 3,538
vegetation images and 21 samples of 6 different species of interest.
Results demonstrate that the proposed change detection approach is
able to effectively identify phenological events.

BibTeX:
@inproceedings{Santos2014ESCIENCE,
  author = {Lilian C. B. dos Santos and Jurandy Almeida and Jefersson A. Santos and Silvio Jamil G. Guimarães and Arnaldo de A. Araújo and Bruna Alberton and Leonor Patrícia Morellato and Ricardo da Silva Torres},
  title = {Phenological event detection by visual rhythm dissimilarity analysis},
  booktitle = {10th IEEE International eScience Conference},
  year = {2014}
}
      


Silva, F.B., Tabbone, S. and da , R. BoG: a New Approach for Graph Matching 2014 22nd International Conference on Pattern Recognition, pp. 82-87  inproceedings  
Abstract: Huge volume of graph data are becoming available. This scenario demands
the development of effective and efficient methods to perform graph
matching. In this paper, we propose to adapt the Bag-of-Words model
into the context of graphs. Using a vocabulary based on graph local
structures, we represent graphs as histograms. Experiments show that
our approach achieves good accuracy rates. Moreover, the advantage
of this representation is that the computation of graph matching
has a very low complexity, which allows to efficiently perform graph
classification and retrieval on large datasets.

BibTeX:
@inproceedings{Silva2014ICPR,
  author = {Fernanda B. Silva and Salvatore Tabbone and Ricardo da},
  title = {BoG: a New Approach for Graph Matching},
  booktitle = {22nd International Conference on Pattern Recognition},
  year = {2014},
  pages = {82-87}
}
      


Almeida, J., Santos, J.A., Alberton, B., Morellato, L.P. and da Silva Torres, R. Plant Species Identification with Phenological Visual Rhythms 2013 eScience (eScience), 2013 IEEE 9th International Conference on, pp. 148-154  inproceedings DOI  
Abstract: Plant phenology studies recurrent plant life cycles events and is
a key component of climate change research. To increase accuracy
of observations, new technologies have been applied for phenological
observation, and one of the most successful are digital cameras,
used as multi-channel imaging sensors to estimate color changes that
are related to phenological events. We monitored leaf-changing patterns
of a cerrado-savanna vegetation by taken daily digital images. We
extract individual plant color information and correlated with leaf
phenological changes. To do so, time series associated with plant
species were obtained, raising the need of using appropriate tools
for mining patterns of interest. In this paper, we present a novel
approach for representing phenological patterns of plant species
derived from digital images. The proposed method is based on encoding
time series as a visual rhythm, which is characterized by image description
algorithms. A comparative analysis of different descriptors is conducted
and discussed. Experimental results show that our approach presents
high accuracy on identifying plant species.

BibTeX:
@inproceedings{Almeida2013ESCIENCE,
  author = {Jurandy Almeida and Jefersson A. Santos and Bruna Alberton and Leonor Patrícia Morellato and Ricardo da Silva Torres},
  title = {Plant Species Identification with Phenological Visual Rhythms},
  booktitle = {eScience (eScience), 2013 IEEE 9th International Conference on},
  year = {2013},
  pages = {148-154},
  doi = {http://doi.org/10.1109/eScience.2013.43}
}
      


Almeida, J., Santos, J.A., Alberton, B., Morellato, L.P. and da S. Torres, R. Visual rhythm-based time series analysis for phenology studies 2013 Image Processing (ICIP), 2013 20th IEEE International Conference on, pp. 4412-4416  inproceedings DOI  
Abstract: Plant phenology has gained importance in the context of global change
research, stimulating the development of new technologies for phenological
observation. In this context, digital cameras have been successfully
used as multi-channel imaging sensors, providing measures to estimate
changes on phenological events, such as leaf flushing and senescence.
We monitored leaf-changing patterns of a cerrado-savanna vegetation
by taken daily digital images. For that, we extract leaf color information
and correlated with phenological changes. In this way, time series
associated with plant species are obtained, raising the need of using
appropriate tools for mining patterns of interest. In this paper,
we present a novel approach for representing phenological patterns
of plant species. The proposed method is based on encoding time series
as a visual rhythm, which is characterized by color description algorithms.
A comparative analysis of different descriptors is conducted and
discussed. Experimental results show that our approach presents high
accuracy on identifying plant species.

BibTeX:
@inproceedings{Almeida2013ICIP,
  author = {Jurandy Almeida and Jefersson A. Santos and Bruna Alberton and Leonor Patrícia Morellato and Ricardo da S. Torres},
  title = {Visual rhythm-based time series analysis for phenology studies},
  booktitle = {Image Processing (ICIP), 2013 20th IEEE International Conference on},
  year = {2013},
  pages = {4412-4416},
  doi = {http://doi.org/10.1109/ICIP.2013.6738909}
}
      


Faria, F.A., Santos, J.A., Sarkar, S., Rocha, A. and da S. Torres, R. Classifier Selection Based on the Correlation of Diversity Measures: When Fewer Is More 2013 Graphics, Patterns and Images (SIBGRAPI), 2013 26th SIBGRAPI - Conference on, pp. 16-23  inproceedings DOI  
Abstract: The ever-growing access to high-resolution images has prompted the
development of region-based classification methods for remote sensing
images. However, in agricultural applications, the recognition of
specific regions is still a challenge as there could be many different
spectral patterns in a same studied area. In this context, depending
on the features used, different learning methods can be used to create
complementary classifiers. Many researchers have developed solutions
based on the use of machine learning techniques to address these
problems. Examples of successful initiatives are those dedicated
to the development of learning techniques for data fusion or Multiple
Classifier Systems (MCS). In MCS, diversity becomes an essential
factor for their success. Different works have been using diversity
measures to select appropriate high-performance classifiers, but
the challenge of finding the optimal number of classifiers for a
target task has not been properly addressed yet. In general, the
proposed solutions rely on the a priori use of ad hoc strategies
for selecting classifiers, followed by the evaluation of their effectiveness
results during training. Searching by the optimal number of classifiers,
however, makes the selection process more expensive. In this paper,
we address this issue by proposing a novel strategy for selecting
classifiers to be combined based on the correlation of different
diversity measures. Diversity measures are used to rank pairs of
classifiers and the agreement among ranked lists guides the classifier
selection process. A fusion framework has been used in our experiments
targeted to the classification of coffee crops in remote sensing
images. Experiment results demonstrate that the novel strategy is
able to yield comparable effectiveness results when contrasted to
several baselines, but using much fewer classifiers.

BibTeX:
@inproceedings{Faria2013SIBGRAPI,
  author = {Fabio A. Faria and Jefersson A. Santos and Sudeep Sarkar and Anderson Rocha and Ricardo da S. Torres},
  title = {Classifier Selection Based on the Correlation of Diversity Measures: When Fewer Is More},
  booktitle = {Graphics, Patterns and Images (SIBGRAPI), 2013 26th SIBGRAPI - Conference on},
  year = {2013},
  pages = {16-23},
  doi = {http://doi.org/10.1109/SIBGRAPI.2013.12}
}
      


Godoi, T.A., da Silva Torres, R., Carvalho, A.M., calves , M.A.G., Ferreira, A.A., Fan, W. and Fox, E.A. A Relevance Feedback Approach for the Author Name Disambiguation Problem 2013 Proceedings of the 13th ACM/IEEE-CS Joint Conference on Digital Libraries, pp. 209-218  inproceedings DOI URL 
Abstract: This paper presents a new name disambiguation method that exploits
user feedback on ambiguous references across iterations. An unsupervised
step is used to define pure training samples, and a hybrid supervised
step is employed to learn a classification model for assigning references
to authors. Our classification scheme combines the Optimum-Path Forest
(OPF) classifier with complex reference similarity functions generated
by a Genetic Programming framework. Experiments demonstrate that
the proposed method yields better results than state-of-the-art disambiguation
methods on two traditional datasets.

BibTeX:
@inproceedings{Godoi2013JCDL,
  author = {Thiago A. Godoi and Ricardo da Silva Torres and Ariadne M.B.R. Carvalho and Marcos A. Gon¸ calves and Anderson A. Ferreira and Weiguo Fan and Edward A. Fox},
  title = {A Relevance Feedback Approach for the Author Name Disambiguation Problem},
  booktitle = {Proceedings of the 13th ACM/IEEE-CS Joint Conference on Digital Libraries},
  publisher = {ACM},
  year = {2013},
  pages = {209--218},
  url = {http://doi.acm.org/10.1145/2467696.2467709},
  doi = {http://doi.org/10.1145/2467696.2467709}
}
      


Muraro, E., Mariano, G., Kozievitch, N.P., Almeida, J., Santos, J.A., da S. Torres, R., Alberton, B. and Morellato, L.P. A Framework for Semantic Annotation of Phenology Image Components 2013 International Conference on Digital Libraries, pp. 243-256  inproceedings  
Abstract: Due to the dissemination of low-cost devices for acquisition, storage,
and sharing, images have been used in several e-science applications.
The use of images in those applications has motivated the creation
of heterogeneous digital objects. In fact, images are not longer
used in isolation and are used to compose other digital objects,
named Complex Objects. In this work, we present a new framework for
automatic semantic annotation of Phenology image components, aiming
at supporting their use in the construction of complex objects. We
propose the application of several approaches for defining appropriate
terms to be used in the annotation process of daily series of Phenology
digital images of plant crowns: ontologies, textual terms, and image
content descriptions. The main contributions of this work are: (i)
the specification of an automatic semantic annotation process for
image components, that takes into account image visual descriptors,
defined textual terms, ontologies, and their combination; and (ii)
the specification and partial implementation of an infrastructure
for annotating image complex objects in the context of Phenology
studies.

BibTeX:
@inproceedings{Muraro2013ICDL,
  author = {Emerson Muraro and Greice Mariano and Nádia P. Kozievitch and Jurandy Almeida and Jefersson A. Santos and Ricardo da S. Torres and Bruna Alberton and Leonor Patrícia Morellato},
  title = {A Framework for Semantic Annotation of Phenology Image Components},
  booktitle = {International Conference on Digital Libraries},
  year = {2013},
  pages = {243-256}
}
      


Pedronette, D.C.G., da Silva Torres, R., Borin, E. and Jr., M.B. Image Re-ranking Acceleration on GPUs 2013 25th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2013, Porto de Galinhas, Pernambuco, Brazil, October 23-26, 2013, pp. 176-183  inproceedings DOI URL 
Abstract: Huge image collections are becoming available lately. In this scenario, the use of Content-Based Image Retrieval (CBIR) systems has emerged as a promising approach to support image searches. The objective of CBIR systems is to retrieve the most similar images in a collection, given a query image, by taking into account image visual properties such as texture, color, and shape. In these systems, the effectiveness of the retrieval process depends heavily on the accuracy of ranking approaches. Recently, re-ranking approaches have been proposed to improve the effectiveness of CBIR systems by taking into account the relationships among images. The re-ranking approaches consider the relationships among all images in a given dataset. These approaches typically demands a huge amount of computational power, which hampers its use in practical situations. On the other hand, these methods can be massively parallelized. In this paper, we propose to speedup the computation of the RL-Sim algorithm, a recently proposed image re-ranking approach, by using the computational power of Graphics Processing Units (GPU). GPUs are emerging as relatively inexpensive parallel processors that are becoming available on a wide range of computer systems. We address the image re-ranking performance challenges by proposing a parallel solution designed to fit the computational model of GPUs. We conducted an experimental evaluation considering different implementations and devices. Experimental results demonstrate that significant performance gains can be obtained. Our approach achieves speedups of 7x from serial implementation considering the overall algorithm and up to 36x on its core steps.

BibTeX:
@inproceedings{Pedronette2013SBAC-PAD,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da Silva Torres and Edson Borin and Mauricio Breternitz Jr.},
  title = {Image Re-ranking Acceleration on GPUs},
  booktitle = {25th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2013, Porto de Galinhas, Pernambuco, Brazil, October 23-26, 2013},
  year = {2013},
  pages = {176--183},
  url = {http://dx.doi.org/10.1109/SBAC-PAD.2013.19},
  doi = {http://doi.org/10.1109/SBAC-PAD.2013.19}
}
      


Pedronette, D.C.G. and da S. Torres, R. Unsupervised Measures for Estimating the Effectiveness of Image Retrieval Systems 2013 Graphics, Patterns and Images (SIBGRAPI), 2013 26th SIBGRAPI - Conference on, pp. 341-348  inproceedings DOI  
Abstract: The main objective of Content-Based Image Retrieval (CBIR) systems
is to retrieve a ranked list containing the most similar images of
a collection given a query image, by taking into account their visual
content. Although these systems represent a very promising approach,
in many situations is very challenging to assure the quality of returned
ranked lists. Supervised approaches rely on training data and information
obtained from user interactions to identify and then improve low-quality
results. However, these approaches require a lot of human efforts
which can be infeasible for many systems. In this paper, we present
two novel unsupervised measures for estimating the effectiveness
of ranked lists in CBIR tasks. Given an estimation of the effectiveness
of ranked lists, many CBIR systems can, for example, emulate the
training process, but now without any user intervention. Improvements
can also be achieved on several unsupervised approaches, such as
re-ranking and rank aggregation methods, once the estimation measures
can help to consider more relevant information by distinguishing
effective from non-effective ranked lists. Both proposed measures
are computed using a novel image representation of ranked lists and
distances among images considering a given dataset. The objective
is to exploit the visual patterns encoded in the image representations
for estimating the effectiveness of ranked lists. Experiments involving
shape, color, and texture descriptors demonstrate that the proposed
approaches can provide accurate estimations of the quality in terms
of effectiveness of ranked lists. The use of proposed measures are
also evaluated in image retrieval tasks aiming at improving the effectiveness
of rank aggregation approaches.

BibTeX:
@inproceedings{Pedronette2013SIBGRAPI,
  author = {Daniel C. G. Pedronette and Ricardo da S. Torres},
  title = {Unsupervised Measures for Estimating the Effectiveness of Image Retrieval Systems},
  booktitle = {Graphics, Patterns and Images (SIBGRAPI), 2013 26th SIBGRAPI - Conference on},
  year = {2013},
  pages = {341-348},
  doi = {http://doi.org/10.1109/SIBGRAPI.2013.54}
}
      


Pereira, L.A.M., Papa, J.P., Almeida, J., da S. Torres, R. and Paraguassu, W.A. A Multiple Labeling-Based Optimum-Path Forest for Video Content Classification 2013 Graphics, Patterns and Images (SIBGRAPI), 2013 26th SIBGRAPI - Conference on, pp. 334-340  inproceedings DOI  
Abstract: Multiple-labeling classification approaches attempt to handle applications
that associate more than one label to a given sample. Since we have
an increasing number of systems that are guided by such assumption,
in this paper we have presented a multiple-labeling approach for
the Optimum-Path Forest (OPF) classifier based on the problem transformation
method. In order to validate our proposal, a multi-labeled video
classification dataset has been used to compare OPF against three
other classifiers and another variant of the OPF classifier based
on a k-neighborhood. The results have shown the validity of the OPF-based
classifiers for multi-labeling classification problems.

BibTeX:
@inproceedings{Pereira2013SIBGRAPI,
  author = {Luis A. M. Pereira and João Paulo Papa and Jurandy Almeida and Ricardo da S. Torres and Willian Amorim Paraguassu},
  title = {A Multiple Labeling-Based Optimum-Path Forest for Video Content Classification},
  booktitle = {Graphics, Patterns and Images (SIBGRAPI), 2013 26th SIBGRAPI - Conference on},
  year = {2013},
  pages = {334-340},
  doi = {http://doi.org/10.1109/SIBGRAPI.2013.53}
}
      


Santos, J.A., Penatti, O.A.B., da Silva Torres, R., Gosselin, P.-H., Philipp-Foliguet, S. and Falcão, A.X. Remote sensing image representation based on hierarchical histogram propagation 2013 Geoscience and Remote Sensing Symposium (IGARSS), 2013 IEEE International, pp. 2982-2985  inproceedings DOI  
Abstract: Many methods have been recently proposed to deal with the large amount
of data provided by high-resolution remote sensing technologies.
Several of these methods rely on the use of image segmentation algorithms
for delineating target objects. However, a common issue in geographic
object-based applications is the definition of the appropriate data
representation scale, a problem that can be addressed by exploiting
multiscale segmentation. The use of multiple scales, however, raises
new challenges related to the definition of effective and efficient
mechanisms for extracting features. In this paper, we address the
problem of extracting histogram-based features from a hierarchy of
regions for multiscale classification. The strategy, called H-Propagation,
exploits the existing relationships among regions in a hierarchy
to iteratively propagate features along multiple scales. The proposed
method speeds up the feature extraction process and yields good results
when compared with global low-level extraction approaches.

BibTeX:
@inproceedings{Santos2013IGARSS,
  author = {Jefersson A. Santos and Otávio A. B. Penatti and Ricardo da Silva Torres and Gosselin, P.-H. and Philipp-Foliguet, S. and Alexandre X. Falcão},
  title = {Remote sensing image representation based on hierarchical histogram propagation},
  booktitle = {Geoscience and Remote Sensing Symposium (IGARSS), 2013 IEEE International},
  year = {2013},
  pages = {2982-2985},
  doi = {http://doi.org/10.1109/IGARSS.2013.6723452}
}
      


Silva, F.B., Goldenstein, S., Tabbone, S. and da Silva Torres, R. Image classification based on bag of visual graphs 2013 Image Processing (ICIP), 2013 20th IEEE International Conference on, pp. 4312-4316  inproceedings DOI  
Abstract: This paper proposes the Bag of Visual Graphs (BoVG), a new approach
to encode the spatial relationships of visual words through a codebook
of visual-word arrangements, represented by graphs. This graph-based
codebook defines a descriptor for image representations that not
only considers the frequency of occurrence of visual words, but also
their spatial relationships. Experiments demonstrate that BoVG yields
high-accuracy scores in classification tasks on the traditional Caltech-101
and Caltech-256 datasets.

BibTeX:
@inproceedings{Silva2013ICIP,
  author = {Fernanda B. Silva and Siome Goldenstein and Salvatore Tabbone and Ricardo da Silva Torres},
  title = {Image classification based on bag of visual graphs},
  booktitle = {Image Processing (ICIP), 2013 20th IEEE International Conference on},
  year = {2013},
  pages = {4312-4316},
  doi = {http://doi.org/10.1109/ICIP.2013.6738888}
}
      


Toffoli, T.O., Kozievitch, N.P., Gonçalves, M.A. and Torres, R.d.S. A Formal Approach for the Specification of Digital Complex Objects 2013 Proceedings of the 19th Brazilian Symposium on Multimedia and the Web, pp. 125-132  inproceedings DOI URL 
Abstract: Complex objects (COs) have surged as a way to integrate different
digital resources under a same logical unit in order to facilitate
aggregation and reuse. However, there is still a lack of consensus
on precise theoretical foundations for COs, especially regarding
design and specification, which compromise their utility and integration
with existing software tools. Moreover, there has been little investigation
on aspects related to the modeling of COs by the end user, much due
to the lack of appropriate tools for this goal. In this work, we
present a new Digital Library (DL) metamodel specially designed for
the CO modeling which is grounded in formal theoretical specification
for COs. More specifically, our goal is two-fold: (i) to indirectly
validate our CO formalization by instantiating it within a DL modeling
tool -- 5SGraph; and (ii) to investigate the difficulties of CO modeling
and specification by real users using the specified metamodel. Experiments
with real users indicate that the use of the metamodel and the graphical
tool facilitates the understanding of the COs structure and the modeling
process.

BibTeX:
@inproceedings{Toffoli2013Webmedia,
  author = {Toffoli, Ticiana Oniki and Kozievitch, Nádia Puchalski and Gonçalves, Marcos André and Torres, Ricardo da Silva},
  title = {A Formal Approach for the Specification of Digital Complex Objects},
  booktitle = {Proceedings of the 19th Brazilian Symposium on Multimedia and the Web},
  publisher = {ACM},
  year = {2013},
  pages = {125--132},
  url = {http://doi.acm.org/10.1145/2526188.2526199},
  doi = {http://doi.org/10.1145/2526188.2526199}
}
      


da S. Torres, R., Hasegawa, M., Tabbone, S., Almeida, J., Santos, J.A., Alberton, B. and Morellato, L.P. Shape-based time series analysis for remote phenology studies 2013 Geoscience and Remote Sensing Symposium (IGARSS), 2013 IEEE International, pp. 3598-3601  inproceedings DOI  
Abstract: Remote phenology has motivated the development of new technologies
for pattern observation. In this scenario, digital cameras have been
used as data source for studies that estimate changes on phenological
events. In this paper, we investigate the use of shape descriptors
in the task of characterizing time series associated with phenological
changes. The main objectives are: i) to determine which color channel
is better for extracting shape descriptors and ii) to analyze the
impact of the sunshine on the performance of shape descriptors.

BibTeX:
@inproceedings{Torres2013IGARSS,
  author = {Ricardo da S. Torres and Makoto Hasegawa and Salvatore Tabbone and Jurandy Almeida and Jefersson A. Santos and Bruna Alberton and Leonor Patrícia Morellato},
  title = {Shape-based time series analysis for remote phenology studies},
  booktitle = {Geoscience and Remote Sensing Symposium (IGARSS), 2013 IEEE International},
  year = {2013},
  pages = {3598-3601},
  doi = {http://doi.org/10.1109/IGARSS.2013.6723608}
}
      


Almeida, J., dos Santos, J.A., Alberton, B., da S. Torres, R. and Morellato, L.P. Remote Phenology: Applying Machine Learning to Detect Phenological Patterns in a Cerrado Savanna 2012 8th IEEE International Conference on E-Science, pp. -  inproceedings DOI  
Abstract: Plant phenology has gained importance in the context of global change
research, stimulating the development of new technologies for phenological
observation. Digital cameras have been successfully used as multi-channel
imaging sensors, providing measures of leaf color change information
(RGB channels), or leafing phenological changes in plants. We monitored
leaf-changing patterns of a cerrado-savanna vegetation by taken daily
digital images. We extract RGB channels from digital images and correlated
with phenological changes. Our first goals were: (1) to test if the
color change information is able to characterize the phenological
pattern of a group of species; and (2) to test if individuals from
the same functional group may be automatically identified using digital
images. In this paper, we present a machine learning approach to
detect phenological patterns in the digital images. Our preliminary
results indicate that: (1) extreme hours (morning and afternoon)
are the best for identifying plant species; and (2) different plant
species present a different behavior with respect to the color change
information. Based on those results, we suggest that individuals
from the same functional group might be identified using digital
images, and introduce a new tool to help phenology experts in the
species identification and location on-the-ground.

BibTeX:
@inproceedings{Almeida2012ESCIENCE,
  author = {Jurandy Almeida and Jefersson A. dos Santos and Bruna Alberton and Ricardo da S. Torres and Leonor Patrícia Morellato},
  title = {Remote Phenology: Applying Machine Learning to Detect Phenological Patterns in a Cerrado Savanna},
  booktitle = {8th IEEE International Conference on E-Science},
  year = {2012},
  pages = {--},
  doi = {http://doi.org/10.1109/eScience.2012.6404438}
}
      


Andrade, F.S.P., Almeida, J., Pedrini, H. and da S. Torres, R. Fusion of Local and Global Descriptors for Content-Based Image and Video Retrieval 2012 17th Iberoamerican Congress on Pattern Recognition, pp. 845-853  inproceedings DOI  
Abstract: Recently, fusion of descriptors has become a trend for improving the
performance in image and video retrieval tasks. Descriptors can be
global or local, depending on how they analyze visual content. Most
of existing works have focused on the fusion of a single type of
descriptor. Different from all of them, this paper aims to analyze
the impact of combining global and local descriptors. Here, we perform
a comparative study of different types of descriptors and all of
their possible combinations. Extensive experiments of a rigorous
experimental design show that global and local descriptors complement
each other, such that, when combined, they outperform other combinations
or single descriptors.

BibTeX:
@inproceedings{Andrade2012CIARP,
  author = {Felipe S. P. Andrade and Jurandy Almeida and Hélio Pedrini and Ricardo da S. Torres},
  title = {Fusion of Local and Global Descriptors for Content-Based Image and Video Retrieval},
  booktitle = {17th Iberoamerican Congress on Pattern Recognition},
  year = {2012},
  pages = {845-853},
  doi = {http://doi.org/10.1007/978-3-642-33275-3%5C_104}
}
      


Bueno, L.M., Valle, E. and da S. Torres, R. Bayesian approach for near-duplicate image detection 2012 International Conference on Multimedia Retrieval, (ICMR), pp. 15  inproceedings DOI  
Abstract: Vote-based algorithms are very popular in tasks based on image local-descriptors,
including object matching, panoramic stitching and near-duplicate
detection. On this paper, we focus on the latter application, proposing
a Bayesian approach, which allows giving a probabilistic interpretation
to the distances between local descriptors in the feature space.
That contrasts with traditional schemes, in which the distances are
used to establish a simple unweighted vote count. Near-duplicate
detection is demanded for a myriad of applications: metadata retrieval
in cultural institutions, detection of copyright violations, duplicate
elimination in storage, etc. The majority of current solutions are
based either on voting algorithms, which are very precise, but expensive;
or on the use of visual dictionaries, which are efficient, but less
precise. Contrarily to raw-vote based systems, our scheme performs
few database accesses; and contrarily to dictionary-based systems,
it allows a fine control of the compromise between precision and
efficiency. In our experiments, it yields 99% accuracy with less
than 10 database accesses, in contrast with the hundreds needed in
raw-voting schemes.

BibTeX:
@inproceedings{Bueno2012ICMR,
  author = {Lucas Moutinho Bueno and Eduardo Valle and Ricardo da S. Torres},
  title = {Bayesian approach for near-duplicate image detection},
  booktitle = {International Conference on Multimedia Retrieval, (ICMR)},
  year = {2012},
  pages = {15},
  doi = {http://doi.org/10.1145/2324796.2324815}
}
      


Faria, F.A., dos Santos, J.A., da S. Torres, R., Rocha, A. and Falcão, A.X. Automatic fusion of region-based classifiers for coffee crop recognition 2012 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 2221-2224  inproceedings DOI  
Abstract: Coffee crop recognition in remote sensing images is a complex task.
It poses several challenges due to different spectral responses and
texture patterns that can be extracted from coffee regions. This
paper presents a novel framework for combining different classifiers
using support vector machine technique (SVM), which try to learn
with each one of classifiers previews experiences (meta-learning).
We investigate the combination of seven learning methods and seven
image descriptors aiming at creating low-cost classifiers for coffee
crops recognition. The objective is to provide an effective mechanism
for coffee crop recognition by fusion of region-based classifiers
in remote sensing images. The experiments showed that the proposed
framework for fusion of classifiers produces better results than
the traditional majority voting fusion approach and all base classifiers
tested.

BibTeX:
@inproceedings{Faria2012IGARSS,
  author = {Fabio A. Faria and Jefersson Alex dos Santos and Ricardo da S. Torres and Anderson Rocha and Alexandre X. Falcão},
  title = {Automatic fusion of region-based classifiers for coffee crop recognition},
  booktitle = {IEEE International Geoscience and Remote Sensing Symposium (IGARSS)},
  year = {2012},
  pages = {2221-2224},
  doi = {http://doi.org/10.1109/IGARSS.2012.6351058}
}
      


Faria, F.A., dos Santos, J.A., Rocha, A. and da S. Torres, R. Automatic Classifier Fusion for Produce Recognition 2012 Conference on Graphics, Patterns and Images (25th SIBGRAPI), pp. 252-259  inproceedings DOI  
Abstract: Recognizing different kinds of fruits and vegetables is a common task
in supermarkets. This task, however, poses several challenges as
it requires the identification of different species of a particular
produce and also its variety. Usually, existing computer-based recognition
approaches are not automatic and demand long-term and laborious prior
training sessions. This paper presents a novel framework for classifier
fusion aiming at supporting the automatic recognition of fruits and
vegetables in a supermarket environment. The objective is to provide
an effective mechanism for combining low-cost classifiers trained
for specific classes of interest. The experiments performed demonstrate
that the proposed framework yields better results than several related
work found in the literature and represents a step forward automatic
produce recognition in cashiers of supermarkets.

BibTeX:
@inproceedings{Faria2012SIBGRAPI,
  author = {Fabio A. Faria and Jefersson A. dos Santos and Anderson Rocha and Ricardo da S. Torres},
  title = {Automatic Classifier Fusion for Produce Recognition},
  booktitle = {Conference on Graphics, Patterns and Images (25th SIBGRAPI)},
  year = {2012},
  pages = {252-259},
  doi = {http://doi.org/10.1109/SIBGRAPI.2012.42}
}
      


Mansano, A.F., Matsuoka, J.A., Afonso, L.C.S., Faria, F.A. and da S. Torres, R. Improving Image Classification Through Descriptor Combination 2012 Conference on Graphics, Patterns and Images (25th SIBGRAPI), pp. 324-329  inproceedings  
Abstract: The efficiency in image classification tasks can be improved using
combined information provided by several sources, such as shape,
color, and texture visual properties. Although many works proposed
to combine different feature vectors, we model the descriptor combination
as an optimization problem to be addressed by evolutionary-based
techniques, which compute distances between samples that maximize
their separability in the feature space. The robustness of the proposed
technique is assessed by the Optimum-Path Forest classifier. Experiments
showed that the proposed methodology can outperform individual information
provided by single descriptors in well-known public datasets.

BibTeX:
@inproceedings{Mansano2012SIBGRAPI,
  author = {Alex Fernandes Mansano and Jéssica Akemi Matsuoka and Luis Cláudio Súgi Afonso and Fabio A. Faria and Ricardo da S. Torres},
  title = {Improving Image Classification Through Descriptor Combination},
  booktitle = {Conference on Graphics, Patterns and Images (25th SIBGRAPI)},
  year = {2012},
  pages = {324-329}
}
      


Mariano, G., Almeida, J., da S. Torres, R., Alberton, B. and Morellato, L.P. Desenvolvimento de um Modelo Conceitual de um Banco de Dados para o Projeto e-phenology 2012 VI e-Science Workshop, XXXII Congresso da Sociedade Brasileira de Computa¸ cão (CSBC), pp. -  inproceedings  
Abstract: Recently, phenology has gained importance as the more simple and reliable
indicator of effects of climate changes on plants and animals. In
this context, we highlight the e-phenology, a multidisciplinary project
combining research in computer science and phenology, in order to
attack theoretical and practical problems involving the use of new
technologies for remote phenological observation. From the computer
science point of view, it is necessary to develop models, tools and
techniques concerning the storage, mining, processing, and retrieving
of information of phenological changes. In this sense, this paper
presents our achievements regarding the modeling and implementation
of a database to manage information handled in the e-phenology project.

BibTeX:
@inproceedings{Mariano2012ESCIENCE-SBC,
  author = {Greice Mariano and Jurandy Almeida and Ricardo da S. Torres and Bruna Alberton and Leonor Patrícia Morellato},
  title = {Desenvolvimento de um Modelo Conceitual de um Banco de Dados para o Projeto e-phenology},
  booktitle = {VI e-Science Workshop, XXXII Congresso da Sociedade Brasileira de Computa¸ cão (CSBC)},
  year = {2012},
  pages = {--}
}
      


Nakamura, R., ao Paulo Papa, J., Fonseca, L.M., dos Santos, J.A. and da S. Torres, R. Hyperspectral Band Selection Through Optimum-Path Forest and Evolutionary-Based Algorithms 2012 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 3066-3069  inproceedings DOI  
Abstract: In this paper we addressed the problem of dimensionality reduction
in hyperspectral imagery classification by combining OPF classifier
together with three recent evolutionary-based optimization algorithms:
PSO, HS and GSA. We conducted experiments with two public datasets
(Indian Pines and Salinas), which demonstrated that OPF combined
with HS and GSA have obtained promising results, being the former
the fastest approach. In regard to Indian Pines dataset, HS and GSA
have achieved close classification rates, but HS has selected 46.25%
less bands, which means a faster feature extraction step. For future
works, we intend to provide a more detailed convergence analysis
for PSO, HS and GSA, and also to introduce novel evolutionary-based
band selection techniques and also to apply these methodologies for
hyperspectral image classification in forest and agriculture applications.

BibTeX:
@inproceedings{Nakamura2012IGARSS,
  author = {Rodrigo Nakamura and João Paulo Papa and Leila M. Fonseca and Jefersson A. dos Santos and Ricardo da S. Torres},
  title = {Hyperspectral Band Selection Through Optimum-Path Forest and Evolutionary-Based Algorithms},
  booktitle = {IEEE International Geoscience and Remote Sensing Symposium (IGARSS)},
  year = {2012},
  pages = {3066-3069},
  doi = {http://doi.org/10.1109/IGARSS.2012.6350778}
}
      


Pedronette, D.C.G. and da S. Torres, R. Combining Re-Ranking and Rank Aggregation Methods 2012 17th Iberoamerican Congress on Pattern Recognition, pp. 170-178  inproceedings DOI  
Abstract: Content-Based Image Retrieval (CBIR) aims at retrieving the most similar
images in a collection by taking into account image visual properties.
In this scenario, accurately ranking collection images is of great
relevance. Aiming at improving the effectiveness of CBIR systems,
re-ranking and rank aggregation algorithms have been proposed. However,
different re-ranking and rank aggregation approaches produce different
image rankings. These rankings are complementary and, therefore,
can be further combined aiming at obtaining more effective results.
This paper presents novel approaches for combining re-ranking and
rank aggregation methods aiming at improving the effectiveness of
CBIR systems. Several experiments were conducted involving shape,
color, and texture descriptors. Experimental results demonstrate
that our approaches can improve the effectiveness of CBIR systems.

BibTeX:
@inproceedings{Pedronette2012CIARP,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da S. Torres},
  title = {Combining Re-Ranking and Rank Aggregation Methods},
  booktitle = {17th Iberoamerican Congress on Pattern Recognition},
  year = {2012},
  pages = {170-178},
  doi = {http://doi.org/10.1007/978-3-642-33275-3%5C_21}
}
      


Pedronette, D.C.G., da Silva Torres, R., Borin, E. and Jr., M.B. Efficient Image Re-Ranking Computation on GPUs 2012 10th IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA), pp. 95-102  inproceedings DOI  
Abstract: The huge growth of image collections and multimedia resources available
is remarkable. One of the most common approaches to support image
searches relies on the use of Content-Based Image Retrieval (CBIR)
systems. CBIR systems aim at retrieving the most similar images in
a collection, given a query image. Since the effectiveness of those
systems is very dependent on the accuracy of ranking approaches,
re-ranking algorithms have been proposed to exploit contextual information
and improve the effectiveness of CBIR systems. Image re-ranking algorithms
typically consider the relationship among every image in a given
dataset when computing the new ranking. This approach demands a huge
amount of computational power, which may render it prohibitive on
very large data sets. In order to mitigate this problem, we propose
using the computational power of Graphics Processing Units (GPU)
to speedup the computation of image re-ranking algorithms. GPUs are
fast emerging and relatively inexpensive parallel processors that
are becoming available on a wide range of computer systems. In this
paper, we propose a parallel implementation of an image re-ranking
algorithm designed to fit the computational model of GPUs. Experimental
results demonstrate that relevant performance gains can be obtained
by our approach.

BibTeX:
@inproceedings{Pedronette2012ISPA,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da Silva Torres and Edson Borin and Mauricio Breternitz Jr.},
  title = {Efficient Image Re-Ranking Computation on GPUs},
  booktitle = {10th IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA)},
  year = {2012},
  pages = {95-102},
  doi = {http://doi.org/10.1109/ISPA.2012.21}
}
      


Penatti, O.A.B., Li, L.T., Almeida, J. and da S. Torres, R. A visual approach for video geocoding using bag-of-scenes 2012 International Conference on Multimedia Retrieval, (ICMR), pp. 53  inproceedings DOI  
Abstract: This paper presents a novel approach for video representation, called
bag-of-scenes. The proposed method is based on dictionaries of scenes,
which provide a high-level representation for videos. Scenes are
elements with much more semantic information than local features,
specially for geotagging videos using visual content. Thus, each
component of the representation model has self-contained semantics
and, hence, it can be directly related to a specific place of interest.
Experiments were conducted in the context of the MediaEval 2011 Placing
Task. The reported results show our strategy compared to those from
other participants that used only visual content to accomplish this
task. Despite our very simple way to generate the visual dictionary,
which has taken photos at random, the results show that our approach
presents high accuracy relative to the state-of-the art solutions.

BibTeX:
@inproceedings{Penatti2012ICMR,
  author = {Otávio Augusto Bizetto Penatti and Lin Tzy Li and Jurandy Almeida and Ricardo da S. Torres},
  title = {A visual approach for video geocoding using bag-of-scenes},
  booktitle = {International Conference on Multimedia Retrieval, (ICMR)},
  year = {2012},
  pages = {53},
  doi = {http://doi.org/10.1145/2324796.2324857}
}
      


dos Santos, J.A., Faria, F., da S. Torres, R., Gosselin, P.-H., Philipp-Foliguet, S., Falcão, A.X. and Rocha, A. Descriptor Correlation Analysis for Remote Sensing Image Multi-Scale Classification 2012 21st International Conference on Pattern Recognition (ICPR)  inproceedings  
Abstract: This paper addresses the problem of remote sensing image multi-scale
classification by: (i) showing that using multiple scales does improve
classification results, but not all scales have the same importance;
(ii) showing that image descriptors do not offer the same contribution
at all scales, as commonly thought, and some of them are very correlated;
(iii) introducing a simple approach to automatically select segmentation
scales, descriptors, and classifiers based on correlation and accuracy
analysis.

BibTeX:
@inproceedings{Santos2012ICPR-1,
  author = {Jefersson Alex dos Santos and Fabio Faria and Ricardo da S. Torres and Philippe-Henri Gosselin and Sylvie Philipp-Foliguet and Alexandre X. Falcão and Anderson Rocha},
  title = {Descriptor Correlation Analysis for Remote Sensing Image Multi-Scale Classification},
  booktitle = {21st International Conference on Pattern Recognition (ICPR)},
  year = {2012}
}
      


dos Santos, J.A., Penatti, O.A., da S. Torres, R., Gosselin, P.-H., Philipp-Foliguet, S. and Falcão, A.X. Improving Texture Description in Remote Sensing Image Multi-Scale Classification Tasks By Using Visual Words 2012 21st International Conference on Pattern Recognition (ICPR)  inproceedings  
Abstract: Although texture features are important for region-based classification
of remote sensing images, the literature shows that texture descriptors
usually have poor performance when compared and combined with color
descriptors. In this paper, we propose a bag-of-visual-words (BOW)
``propagation" approach to extract texture features from a hierarchy
of regions. This strategy improves efficacy of feature as it encodes
texture information independently of the region shape. Experiments
show that the proposed approach improves the classification results
when compared with global descriptors using the bounding box padding
strategy.

BibTeX:
@inproceedings{Santos2012ICPR-2,
  author = {Jefersson Alex dos Santos and Otávio A.B. Penatti and Ricardo da S. Torres and Philippe-Henri Gosselin and Sylvie Philipp-Foliguet and Alexandre X. Falcão},
  title = {Improving Texture Description in Remote Sensing Image Multi-Scale Classification Tasks By Using Visual Words},
  booktitle = {21st International Conference on Pattern Recognition (ICPR)},
  year = {2012}
}
      


Vidal, M., ao M. B. Cvalcanti, J., Moura, E.S., Silva, A.S. and da S. Torres, R. Sorted Dominant Local Color for Searching Large and Heterogeneous Image Databases 2012 21st International Conference on Pattern Recognition (ICPR)  inproceedings  
Abstract: Recent work on Content-Based Image Retrieval (CBIR) have presented
alternative methods for fast image indexing and retrieval using Bags
of Visual Words (BoVW). In such methods, images are represented as
sets of visual words, which can be indexed and searched using well-known
text retrieval techniques, allowing fast search on large image databases.
In this paper we propose a novel method based on BoVW that improves
over current methods by using a new kind of local color descriptor,
which we call SDLC, that encodes the most predominant color occurrences
in blocks of different image regions. We report results of experiments
we performed with two publicly available image databases. The results
indicate that the use of SDLC led to a quite competitive CBIR method
in comparison to the state-of-the-art.

BibTeX:
@inproceedings{Vidal2012ICPR,
  author = {Márcio Vidal and João M. B. Cvalcanti and Edleno S. Moura and Altigran S. Silva and Ricardo da S. Torres},
  title = {Sorted Dominant Local Color for Searching Large and Heterogeneous Image Databases},
  booktitle = {21st International Conference on Pattern Recognition (ICPR)},
  year = {2012}
}
      


Almeida, J., Leite, N.J. and da S. Torres, R. Rapid Cut Detection on Compressed Video 2011 16th Iberoamerican Congress (CIARP), pp. 71-78  inproceedings DOI  
Abstract: The temporal segmentation of a video sequence is one of the most important
aspects for video processing, analysis, indexing, and retrieval.
Most of existing techniques to address the problem of identifying
the boundary between consecutive shots have focused on the uncompressed
domain. However, decoding and analyzing of a video sequence are two
extremely time-consuming tasks. Since video data are usually available
in compressed form, it is desirable to directly process video material
without decoding. In this paper, we present a novel approach for
video cut detection that works in the compressed domain. The proposed
method is based on both exploiting visual features extracted from
the video stream and on using a simple and fast algorithm to detect
the video transitions. Experiments on a real-world video dataset
with several genres show that our approach presents high accuracy
relative to the state-of-the-art solutions and in a computational
time that makes it suitable for online usage.

BibTeX:
@inproceedings{Almeida2011CIARP,
  author = {Jurandy Almeida and Neucimar J. Leite and Ricardo da S. Torres},
  title = {Rapid Cut Detection on Compressed Video},
  booktitle = {16th Iberoamerican Congress (CIARP)},
  year = {2011},
  pages = {71-78},
  doi = {http://doi.org/10.1007/978-3-642-25085-9%5C_8}
}
      


Almeida, J., Leite, N.A. and da S. Torres, R. Comparison of Video Sequences with Histograms of Motion Patterns 2011 IEEE International Conference on Image Processing, pp. 3673-3676  inproceedings DOI  
Abstract: Making efficient use of video information requires the development
of a video signature and a similarity measure to rapidly identify
similar videos in a huge database. Most of existing techniques to
address this problem have focused on the uncompressed domain. However,
decoding and analyzing of a video sequence are extremely time-consuming
tasks. Since video data are usually available in compressed form,
it is desirable to directly process video material without decoding.
In this paper, we present a novel approach for comparing video sequences
that works in the compressed domain. The proposed method is based
on recognizing motion patterns extracted from the video stream and
their occurrence histogram is proven to be a powerful feature for
describing the video content. Experiments on a TRECVID 2010 dataset
show that our approach presents high accuracy relative to the state-of-
the-art solutions and in a computational time that makes it suitable
for large collections.

BibTeX:
@inproceedings{Almeida2011ICIP,
  author = {Jurandy Almeida and Neucimar A. Leite and Ricardo da S. Torres},
  title = {Comparison of Video Sequences with Histograms of Motion Patterns},
  booktitle = {IEEE International Conference on Image Processing},
  year = {2011},
  pages = {3673-3676},
  doi = {http://doi.org/10.1109/ICIP.2011.6116516}
}
      


Kozievitch, N.P., da S. Torres, R., Santanchè, A. and Leite, N.J. Reusing a compound-based infrastructure for searching video stories 2011 IEEE International Conference on Information Reuse and Integration (IRI), pp. 222-227  inproceedings DOI  
Abstract: The fast evolution of technology has led to a growing demand for multimedia
data, increasing the amount of research into efficient systems to
manage those materials. A lot of research has being done by the Content-Based
Image Retrieval (CBIR) community in the field of images. Nowadays,
they play a key role in digital applications. Thus, contextual integration
of images with different sources is vital. It involves reusing and
aggregating a large amount of information with other media types.
In particular, if we consider video data, images can be used to summarize
videos into storyboards, providing an easy way to navigate and to
browse large video collections. This has been the goal of a quickly
evolving research area known as video summarization. In this paper,
we present a novel approach to reuse the CBIR infrastructure for
searching video stories, taking advantage of the compound object
(CO) concept to integrate resources. Our approach relies on a specific
component technology to encapsulate the CBIR related tasks and integrate
them with video summarization techniques, known as Digital Content
Component (DCC). Such a strategy provides an effective way to reuse,
compose, and aggregate both content and processing software.

BibTeX:
@inproceedings{Kozievitch2011IRI,
  author = {Nádia P. Kozievitch and Ricardo da S. Torres and André Santanchè and Neucimar J. Leite},
  title = {Reusing a compound-based infrastructure for searching video stories},
  booktitle = {IEEE International Conference on Information Reuse and Integration (IRI)},
  year = {2011},
  pages = {222--227},
  doi = {http://doi.org/10.1109/IRI.2011.6009550}
}
      



24th SIBGRAPI Conference on Graphics, Patterns and Images, Sibgrapi 2011, Alagoas, Maceió, Brazil, August 28-31, 2011 2011   proceedings URL 
BibTeX:
@proceedings{Lewiner2011SIBGRAPI,,
  title = {24th SIBGRAPI Conference on Graphics, Patterns and Images, Sibgrapi 2011, Alagoas, Maceió, Brazil, August 28-31, 2011},
  publisher = {IEEE},
  year = {2011},
  url = {http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=6134712}
}
      


Murthy, U., Li, L.T., Hallerman, E., Fox, E.A., Pérez-Quiñones, M.A., Delcambre, L.M.L. and da S. Torres, R. Use of subimages in fish species identification: a qualitative study 2011 Joint International Conference on Digital Libraries (JCDL), pp. 185-194  inproceedings DOI  
Abstract: Many scholarly tasks involve working with subdocuments, or contextualized
fine-grain information, i.e., with information that is part of some
larger unit. A digital library (DL) facilitates management, access,
retrieval, and use of collections of data and metadata through services.
However, most DLs do not provide infrastructure or services to support
working with subdocuments. Superimposed information (SI) refers to
new information that is created to reference subdocuments in existing
information resources. We combine this idea of SI with traditional
DL services, to define and develop a DL with SI (SI-DL). We explored
the use of subimages and evaluated the use of SuperIDR, a prototype
SI-DL, in fish species identification, a scholarly task that involves
working with subimages. The contexts and strategies of working with
subimages in SuperIDR suggest new and enhanced support (SI-DL services)
for scholarly tasks that involve working with subimages, including
new ways of querying and searching for subimages and associated information.
The main conceptual contributions of our work are the insights gained
from these findings of the use of subimages and of SuperIDR, which
lead to recommendations for the design of digital libraries with
superimposed information.

BibTeX:
@inproceedings{Murthy2011JCDL,
  author = {Uma Murthy and Lin Tzy Li and Eric Hallerman and Edward A. Fox and Manuel A. Pérez-Quiñones and Lois M. L. Delcambre and Ricardo da S. Torres},
  title = {Use of subimages in fish species identification: a qualitative study},
  booktitle = {Joint International Conference on Digital Libraries (JCDL)},
  year = {2011},
  pages = {185--194},
  doi = {http://doi.org/10.1145/1998076.1998112}
}
      


Pedronette, D.C.G. and da S. Torres, R. Image Re-ranking and Rank Aggregation Based on Similarity of Ranked Lists 2011 14th International Conference on Computer Analysis of Images and Patterns (CAIP), pp. 369-376  inproceedings DOI  
Abstract: The objective of Content-based Image Retrieval (CBIR) systems is to
return a ranked list containing the most similar images in a collection
given a query image. The effectiveness of these systems is very dependent
on the accuracy of the distance function adopted. In this paper,
we present a novel approach for redefining distances and later re-ranking
images aiming to improve the effectiveness of CBIR systems. In our
approach, distance among images are redefined based on the similarity
of their ranked lists. Conducted experiments involving shape, color,
and texture descriptors demonstrate the effectiveness of our method

BibTeX:
@inproceedings{Pedronette2011CAIP,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da S. Torres},
  title = {Image Re-ranking and Rank Aggregation Based on Similarity of Ranked Lists},
  booktitle = {14th International Conference on Computer Analysis of Images and Patterns (CAIP)},
  year = {2011},
  pages = {369--376},
  doi = {http://doi.org/10.1007/978-3-642-23672-3%5C_45}
}
      


Pedronette, D.C.G. and da S. Torres, R. Exploiting Contextual Information for Rank Aggregation 2011 IEEE International Conference on Image Processing, pp. 97-100  inproceedings DOI  
Abstract: This paper presents a novel rank aggregation approach based on contextual
information aiming to improve the effectiveness of Content-Based
Image Retrieval (CBIR) tasks. In our approach, information encoded
in both distances among images and ranked lists computed by CBIR
systems are used for analyzing contextual information and then re-rank
collection images. We conducted several experiments involving shape,
color, and texture descriptors. We also evaluated our method in comparison
to other rank aggregation approaches. Experimental results demonstrate
the effectiveness of our method.

BibTeX:
@inproceedings{Pedronette2011ICIP,
  author = {Daniel C. G. Pedronette and Ricardo da S. Torres},
  title = {Exploiting Contextual Information for Rank Aggregation},
  booktitle = {IEEE International Conference on Image Processing},
  year = {2011},
  pages = {97--100},
  doi = {http://doi.org/10.1109/ICIP.2011.6116726}
}
      


Pedronette, D.C.G. and da S. Torres, R. Exploiting Contextual Spaces for Image Re-Ranking and Rank Aggregation 2011 Multimedia Information Retrieval, pp. 13:1 - 13:8  inproceedings DOI  
Abstract: The objective of Content-based Image Retrieval (CBIR) systems is to
return the most similar images given an image query. In this scenario,
accurately ranking collection images is of great relevance. In general,
CBIR systems consider only pairwise image analysis, that is, compute
similarity measures considering only pair of images, ignoring the
rich information encoded in the relations among several images. This
paper presents a novel re-ranking approach based on contextual spaces
aiming to improve the effectiveness of CBIR tasks, by exploring relations
among images. In our approach, information encoded in both distances
among images and ranked lists computed by CBIR systems are used for
analyzing contextual information. The re-ranking method can also
be applied to other tasks, such as: (i) for combining ranked lists
obtained by using different image descriptors (rank aggregation);
and (ii) for combining post-processing methods. We conducted several
experiments involving shape, color, and texture descriptors and comparisons
to other post-processing methods. Experimental results demonstrate
the effectiveness of our method.

BibTeX:
@inproceedings{Pedronette2011ICMR,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da S. Torres},
  title = {Exploiting Contextual Spaces for Image Re-Ranking and Rank Aggregation},
  booktitle = {Multimedia Information Retrieval},
  year = {2011},
  pages = {13:1 -- 13:8},
  doi = {http://doi.org/10.1145/1991996.1992009}
}
      


Penatti, O.A.B., Valle, E. and da S. Torres, R. Encoding Spatial Arrangement of Visual Words 2011 16th Iberoamerican Congress (CIARP), pp. 240-247  inproceedings DOI  
Abstract: This paper presents a new approach to encode spatial-relationship
information of visual words in the well-known visual dictionary model.
The current most popular approach to describe images based on visual
words is by means of bags-of-words which do not encode any spatial
information. We propose a graceful way to capture spatial-relationship
information of visual words that encodes the spatial arrangement
of every visual word in an image. Our experiments show the importance
of the spatial information of visual words for image classification
and show the gain in classification accuracy when using the new method.
The proposed approach creates opportunities for further improvements
in image description under the visual dictionary model.

BibTeX:
@inproceedings{Penatti2011CIARP,
  author = {Otávio Augusto Bizetto Penatti and Eduardo Valle and Ricardo da S. Torres},
  title = {Encoding Spatial Arrangement of Visual Words},
  booktitle = {16th Iberoamerican Congress (CIARP)},
  year = {2011},
  pages = {240--247},
  doi = {http://doi.org/10.1007/978-3-642-25085-9%5C_28}
}
      


dos Santos, J.A., da Silva, A.T., da S. Torres, R., Falcão, A.X., Magalhães, L.P. and Lamparelli, R.A.C. Interactive Classification of Remote Sensing Images by Using Optimum-Path Forest and Genetic Programming 2011 14th International Conference on Computer Analysis of Images and Patterns (CAIP), pp. 300-307  inproceedings DOI  
Abstract: The use of remote sensing images as a source of information in agribusiness
applications is very common. In those applications, it is fundamental
to know how the space occupation is. However, identification and
recognition of crop regions in remote sensing images are not trivial
tasks yet. Although there are automatic methods proposed to that,
users very often prefer to identify regions manually. That happens
because these methods are usually developed to solve specific problems,
or, when they are of general purpose, they do not yield satisfying
results. This work presents a new interactive approach based on relevance
feedback to recognize regions of remote sensing. Relevance feedback
is a technique used in content-based image retrieval (CBIR) tasks.
Its objective is to aggregate user preferences to the search process.
The proposed solution combines the Optimum-Path Forest (OPF) classifier
with composite descriptors obtained by a Genetic Programming (GP)
framework. The new approach has presented good results with respect
to the identification of pasture and coffee crops, overcoming the
results obtained by a recently proposed method and the traditional
Maximimun Likelihood algorithm.

BibTeX:
@inproceedings{Santos2011CAIP,
  author = {Jefersson Alex dos Santos and André Tavares da Silva and Ricardo da S. Torres and Alexandre X. Falcão and Léo Pini Magalhães and Rubens A. C. Lamparelli},
  title = {Interactive Classification of Remote Sensing Images by Using Optimum-Path Forest and Genetic Programming},
  booktitle = {14th International Conference on Computer Analysis of Images and Patterns (CAIP)},
  year = {2011},
  pages = {300--307},
  doi = {http://doi.org/10.1007/978-3-642-23678-5%5C_35}
}
      


Teodoro, G., Valle, E., Mariano, N., da S. Torres, R. and Jr., W.M. Adaptive parallel approximate similarity search for responsive multimedia retrieval 2011 20th ACM Conference on Information and Knowledge Management (CIKM), pp. 495-504  inproceedings DOI  
Abstract: This paper introduces Hypercurves, a flexible framework for providing
similarity search indexing to high throughput multimedia services.
Hypercurves efficiently and effectively answers k-nearest neighbor
searches on multigigabyte high-dimensional databases. It supports
massively parallel processing and adapts at runtime its parallelization
regimens to keep answer times optimal for either low and high demands.
In order to achieve its goals, Hypercurves introduces new techniques
for selecting parallelism configurations and allocating threads to
computation cores, including hyperthreaded cores. Its efficiency
gains are throughly validated on a large database of multimedia descriptors,
where it presented near linear speedups and superlinear scaleups.
The adaptation reduces query response times in 43% and 74% for
both platforms tested, when compared to the best static parallelism
regimens.

BibTeX:
@inproceedings{Teodoro2011CIKM,
  author = {George Teodoro and Eduardo Valle and Nathan Mariano and Ricardo da S. Torres and Wagner Meira Jr.},
  title = {Adaptive parallel approximate similarity search for responsive multimedia retrieval},
  booktitle = {20th ACM Conference on Information and Knowledge Management (CIKM)},
  year = {2011},
  pages = {495--504},
  doi = {http://doi.org/10.1145/2063576.2063651}
}
      


Akune, F., Valle, E. and da S. Torres, R. MONORAIL: A Disk-Friendly Index for Huge Descriptor Databases 2010 20th International Conference on Pattern Recognition, (ICPR), pp. 4145-4148  inproceedings DOI  
Abstract: We propose MONORAIL, an indexing scheme for very large multimedia
descriptor databases. Our index is based on the Hilbert curve, which
is able to map the high-dimensional space of those descriptors to
a single dimension. Instead of using several curves to mitigate boundary
effects, we use a single curve with several surrogate points for
each descriptor. Thus, we are able to reduce the random accesses
to the bare minimum. In a rigorous empirical comparison with another
method based on multiple surrogates, ours shows a significant improvement,
due to our careful choice of the surrogate points.

BibTeX:
@inproceedings{Akune2010ICPR,
  author = {Fernando Akune and Eduardo Valle and Ricardo da S. Torres},
  title = {MONORAIL: A Disk-Friendly Index for Huge Descriptor Databases},
  booktitle = {20th International Conference on Pattern Recognition, (ICPR)},
  year = {2010},
  pages = {4145--4148},
  doi = {http://doi.org/10.1109/ICPR.2010.1008}
}
      


Almeida, J., da S. Torres, R. and Leite, N.J. Rapid Video Summarization on Compressed Video 2010 IEEE International Symposium on Multimedia (ISM), pp. 113-120  inproceedings DOI  
Abstract: Recent advances in technology have increased the availability of video
data, creating a strong requirement for efficient systems to manage
those materials. Making efficient use of video information requires
that data be accessed in a user-friendly way. This has been the goal
of a quickly evolving research area known as video summarization.
Most of existing techniques to address the problem of summarizing
a video sequence have focused on the uncompressed domain. However,
decoding and analyzing of a video sequence are two extremely time-consuming
tasks. Since video data are usually available in compressed form,
it is desirable to directly process video material without decoding.
In this paper, we present a novel approach for video summarization
that works in the compressed domain. The proposed method is based
on both exploiting visual features extracted from the video stream
and on using a simple and fast algorithm to summarize the video content.
Results from a rigorous empirical comparison with a subjective evaluation
show that our approach produces video summaries with superior quality
relative to the state-of-the-art solutions and in a computational
time that allows on-the-fly usage.

BibTeX:
@inproceedings{Almeida2010ISM,
  author = {Jurandy Almeida and Ricardo da S. Torres and Neucimar J. Leite},
  title = {Rapid Video Summarization on Compressed Video},
  booktitle = {IEEE International Symposium on Multimedia (ISM)},
  year = {2010},
  pages = {113--120},
  doi = {http://doi.org/10.1109/ISM.2010.25}
}
      


Almeida, J., Minetto, R., Almeida, T., da S. Torres, R. and Leite, N.J. Estimation of Camera Parameters in Videos Sequence with a Large Amount of Scene Motion 2010 17th International Conference on Systems, Signals, and Image Processing (IWSSIP), pp. 348-351  inproceedings  
Abstract: Most of existing techniques to estimate camera motion is based on
analysis of the optical flow. However, such methods can be inaccurate
and/or inefficiently when applied in video sequences which have a
large amount of motion or a large number of scene changes. In this
paper, we present an approach to estimate camera motion based on
analysis of local invariant features. Such features are robust across
a substantial range of affine distortion. Experiments on synthesized
video clips with a fully controlled environment show that our technique
is more effective than the optical flow-based approaches for estimating
camera motion with a large amount of scene motion.

BibTeX:
@inproceedings{Almeida2010IWSSIP,
  author = {Jurandy Almeida and Rodrigo Minetto and Tiago Almeida and Ricardo da S. Torres and Neucimar J. Leite},
  title = {Estimation of Camera Parameters in Videos Sequence with a Large Amount of Scene Motion},
  booktitle = {17th International Conference on Systems, Signals, and Image Processing (IWSSIP)},
  year = {2010},
  pages = {348--351}
}
      


Faria, F.A., Papa, J.P., da S. Torres, R. and Falcão, A.X. Multimodal Pattern Recognition Through Particle Swarm Optimization 2010 17th International Conference on Systems, Signals, and Image Processing (IWSSIP), pp. 134-137  inproceedings  
Abstract: In this paper we introduce the idea of descriptor combination by Particle
Swarm Optimization and its applications for classification purposes
using a recently pattern recognition technique called Optimum-Path
Forest (OPF), which interprets the samples of the dataset as the
nodes of a given graph, and each arc is weighted by the distance
between the corresponding nodes. The method combines different color
descriptors for classification in the COREL dataset to further weight
the OPF graph arcs by these combined similarities. We show that our
proposed approach can provides better accuracies than the individual
ones obtained by each descriptor.

BibTeX:
@inproceedings{Faria2010IWSSIP,
  author = {Fabio A. Faria and João P. Papa and Ricardo da S. Torres and Alexandre X. Falcão},
  title = {Multimodal Pattern Recognition Through Particle Swarm Optimization},
  booktitle = {17th International Conference on Systems, Signals, and Image Processing (IWSSIP)},
  year = {2010},
  pages = {134--137}
}
      


Faria, F.A., Veloso, A., de Almeida, H.M., Valle, E., da S. Torres, R., Gonçalves, M.A. and Jr., W.M. Learning to rank for content-based image retrieval 2010 Multimedia Information Retrieval (MIR), pp. 285-294  inproceedings DOI  
Abstract: In Content-based Image Retrieval (CBIR), accurately ranking the returned
images is of paramount importance, since users consider mostly the
topmost results. The typical ranking strategy used by many CBIR systems
is to employ image content descriptors, so that returned images that
are most similar to the query image are placed higher in the rank.
While this strategy is well accepted and widely used, improved results
may be obtained by combining multiple image descriptors. In this
paper we explore this idea, and introduce algorithms that learn to
combine information coming from different descriptors. The proposed
learning to rank algorithms are based on three diverse learning techniques:
Support Vector Machines (CBIR-SVM), Genetic Programming (CBIR-GP),
and Association Rules (CBIR-AR). Eighteen image content descriptors(color,
texture, and shape information) are used as input and provided as
training to the learning algorithms. We performed a systematic evaluation
involving two complex and heterogeneous image databases (Corel e
Caltech) and two evaluation measures (Precision and MAP). The empirical
results show that all learning algorithms provide significant gains
when compared to the typical ranking strategy in which descriptors
are used in isolation. We concluded that, in general, CBIR-AR and
CBIR-GP outperforms CBIR-SVM. A fine-grained analysis revealed the
lack of correlation between the results provided by CBIR-AR and the
results provided by the other two algorithms, which indicates the
opportunity of an advantageous hybrid approach.

BibTeX:
@inproceedings{Faria2010MIR,
  author = {Fabio Augusto Faria and Adriano Veloso and Humberto Mossri de Almeida and Eduardo Valle and Ricardo da S. Torres and Marcos André Gonçalves and Wagner Meira Jr.},
  title = {Learning to rank for content-based image retrieval},
  booktitle = {Multimedia Information Retrieval (MIR)},
  year = {2010},
  pages = {285--294},
  doi = {http://doi.org/10.1145/1743384.1743434}
}
      


Garcia, V.B. and da S. Torres, R. A Shape Descriptor based on Scale-invariant Multiscale Fractal Dimension 2010
Vol. 25th International Conference on Computer Vision Theory and Applications, pp. 185-190 
inproceedings  
Abstract: This paper proposes a new scale-invariant shape descriptor based on
the Multiscale Fractal Dimension (MFD). The MFD is a curve that describes
boundary complexity and self-affinity characteristics by obtaining
fractal dimension values as function of Euclidean morphologic dilation
radii. Using this concept, which guarantees rotation and translation
invariance, we introduce a new scale-invariant descriptor that is
obtained by selecting a relevant fragment of this curve using a sliding
window. The novel shape descriptor is compared with the Multiscale
Fractal Dimension and four other shape descriptors. Experimental
results demonstrate that the new descriptor is scale-invariant and
yields very good results in terms of effectiveness performace when
compared with well-known shape descriptors.

BibTeX:
@inproceedings{Garcia2010VISAPP,
  author = {Vítor Baccetti Garcia and Ricardo da S. Torres},
  title = {A Shape Descriptor based on Scale-invariant Multiscale Fractal Dimension},
  booktitle = {5th International Conference on Computer Vision Theory and Applications},
  year = {2010},
  volume = {2},
  pages = {185--190}
}
      


Gil, F.B., Kozievitch, N.P. and da S. Torres, R. A Geographic Annotation Service for Biodiversity Systems 2010 XI Brazilian Symposium on Geoinformatics (Geoinfo), pp. 33-44  inproceedings  
Abstract: Biodiversity studies are often based on the use of data associated
with field observations. These data are usually associated with a
geographic location. Most of existing biodiversity information systems
provides support for storing and querying geographic data. Annotation
services, in general, are not supported. This paper presents an annotation
Web service to correlate biodiversity data and geographic information.
We use superimposed information concepts for constructing a Web service
for annotating vector geographic data. The Web service specification
includes the definition of a generic API for handling annotations
and the definition of a data model for storing them. The solution
was validated through the implementation of a prototype for the biodiversity
area considering a potential usage scenario.

BibTeX:
@inproceedings{Gil2010Geoinfo,
  author = {Fabiana B. Gil and Nádia P. Kozievitch and Ricardo da S. Torres},
  title = {A Geographic Annotation Service for Biodiversity Systems},
  booktitle = {XI Brazilian Symposium on Geoinformatics (Geoinfo)},
  year = {2010},
  pages = {33--44}
}
      


Kozievitch, N.P., da S. Torres, R., Park, S.H., Fox, E.A., Short, N., Abbott, L., Misra, S. and Hsiao, M. Rethinking Fingerprint Evidence Through Integration of Very Large Digital Libraries 2010 Third Workshop on Very Large Digital Libraries, 14th European Conference on Research and Advanced Technologies on Digital Libraries  inproceedings  
Abstract: Fingerprints play a key role in biometrics and forensic sci- ence
because of their uniqueness. Essential is contextual integration
of fingerprint evidence from different sources, which involves composing,
reusing, and aggregating a large amount of information. Thus, this
pa- per (1) describes different types of fingerprint information
from a digital library perspective; (2) investigates compound object
concepts as used in connection with fingerprints; and (3) presents
a preliminary integration of very large fingerprint digital libraries.

BibTeX:
@inproceedings{Kozievitch2010WVLDL-ECDL,
  author = {Nádia P. Kozievitch and Ricardo da S. Torres and Sung Hee Park and Edward A. Fox and Nathan Short and Lynn Abbott and Supratik Misra and Michael Hsiao},
  title = {Rethinking Fingerprint Evidence Through Integration of Very Large Digital Libraries},
  booktitle = {Third Workshop on Very Large Digital Libraries, 14th European Conference on Research and Advanced Technologies on Digital Libraries},
  year = {2010}
}
      


Macário, C.G.N., dos Santos, J.A., Medeiros, C.B. and da S. Torres, R. Annotating data to support decision-making: a case study 2010 6th Workshop on Geographic Information Retrieval, (GIR)  inproceedings DOI  
Abstract: Georeferenced data are a key factor in many decision-making systems.
However, their interpretation is user and context dependent so that,
for each situation, data analysts have to interpret them, a time-consuming
task. One approach to alleviate this task, is the use of semantic
annotations to store the produced information. Annotating data is
however hard to perform and prone to errors, especially when executed
manually. This difficulty increases with the amount of data to annotate.
Moreover, annotation requires multi-disciplinary collaboration of
researchers, with access to heterogeneous and distributed data sources
and scientific computations. This paper illustrates our solution
to approach this problem by means of a case study in agriculture.
It shows how our implementation of a framework to automate the annotation
of geospatial data can be used to process real data from remote sensing
images and other official Brazilian data sources.

BibTeX:
@inproceedings{Macario2010GIR,
  author = {Carla Geovana N. Macário and Jefersson Alex dos Santos and Claudia Bauzer Medeiros and Ricardo da S. Torres},
  title = {Annotating data to support decision-making: a case study},
  booktitle = {6th Workshop on Geographic Information Retrieval, (GIR)},
  year = {2010},
  doi = {http://doi.org/10.1145/1722080.1722106}
}
      


Murthy, U., Kozievitch, N.P., Fox, E.A., da S. Torres, R. and Hallerm, E.M. SUPERIDR: A Tablet PC Tool for Image Description and Retrieval 2010 Workshop on the Impact of Pen-Based Technology on Education (WIPTE)  inproceedings  
Abstract: SuperIDR is a tablet-PC-based superimposed image description and retrieval
tool, which combines text and visual content-based image description
and retrieval. It allows users to select and mark parts of images
and associate them with text annotations. Annotations can be entered
using either pen-based or keypad-based input. Later, users can retrieve
information through text- and content-based search on textual descriptions,
annotations, images, and parts of images. We developed SuperIDR as
an aid to species identification for the fisheries domain, working
with freshwater fishes of Virginia. We adapted the tool to work with
parasite images and descriptions. We tested both versions in fisheries
and biology classes at Virginia Tech and UNICAMP (Brazil) and found
the tool to be a useful aid to species identification. In this paper,
we provide an overview of SuperIDR and the evaluations that we conducted.

BibTeX:
@inproceedings{Murthy2010WIPTE,
  author = {Uma Murthy and Nádia P. Kozievitch and Edward. A. Fox and Ricardo da S. Torres and Eric M. Hallerm},
  title = {SUPERIDR: A Tablet PC Tool for Image Description and Retrieval},
  booktitle = {Workshop on the Impact of Pen-Based Technology on Education (WIPTE)},
  year = {2010}
}
      


Pedronette, D.C.G. and da S. Torres, R. Exploiting Contextual Information for Image Re-ranking 2010 15th Iberoamerican Congress on Pattern Recognition, (CIARP), pp. 541-548  inproceedings DOI  
Abstract: This paper presents a novel re-ranking approach based on contextual
information used to improve the effectiveness of Content-Based Image
Retrieval (CBIR) tasks. In our approach, image processing techniques
are applied to ranked lists defined by CBIR descriptors. Conducted
experiments involving shape, color, and texture descriptors demonstrate
the effectiveness of our method.

BibTeX:
@inproceedings{Pedronette2010CIARP,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da S. Torres},
  title = {Exploiting Contextual Information for Image Re-ranking},
  booktitle = {15th Iberoamerican Congress on Pattern Recognition, (CIARP)},
  year = {2010},
  pages = {541--548},
  doi = {http://doi.org/10.1007/978-3-642-16687-7%5C_71}
}
      


Pedronette, D.C.G. and da S. Torres, R. Distances Correlation for Re-Ranking in Content-Based Image Retrieval 2010 23rd SIBGRAPI Conference on Graphics, Patterns and Images, pp. 1-8  inproceedings DOI  
Abstract: Content-based image retrieval relies on the use of efficient and effective
image descriptors. One of the most important components of an image
descriptor is concerned with the distance function used to measure
how similar two images are. This paper presents a clustering approach
based on distances correlation for computing the similarity among
images. Conducted experiments involving shape, color, and texture
descriptors demonstrate the effectiveness of our method.

BibTeX:
@inproceedings{Pedronette2010SIBGRAPI,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da S. Torres},
  title = {Distances Correlation for Re-Ranking in Content-Based Image Retrieval},
  booktitle = {23rd SIBGRAPI Conference on Graphics, Patterns and Images},
  year = {2010},
  pages = {1--8},
  doi = {http://doi.org/10.1109/SIBGRAPI.2010.9}
}
      


Pedronette, D.C.G. and da S. Torres, R. Shape Retrieval using Contour Features and Distance Optimization 2010
Vol. 25th International Conference on Computer Vision Theory and Applications, pp. 197-202 
inproceedings  
Abstract: This paper presents a shape descriptor based on a set of features
computed for each point of an object contour. We also present an
algorithm for distance optimization based on the similarity among
ranked lists. Experiments were conducted on two well-known data sets:
MPEG-7 and Kimia. Experimental results demonstrate that the combination
of the two methods is very effective and yields better results than
recently proposed shape descriptors.

BibTeX:
@inproceedings{Pedronette2010VISAPP,
  author = {Daniel Carlos Guimarães Pedronette and Ricardo da S. Torres},
  title = {Shape Retrieval using Contour Features and Distance Optimization},
  booktitle = {5th International Conference on Computer Vision Theory and Applications},
  year = {2010},
  volume = {2},
  pages = {197--202}
}
      


dos Santos, J.A., Faria, F.A., Calumby, R.T., da Silva Torres, R. and Lamparelli, R.A.C. A Genetic Programming approach for coffee crop recognition 2010 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 3418-3421  inproceedings DOI  
Abstract: This work presents a new approach for automatic recognition of coffee
crops in RSIs. The method applies an approach based on Genetic Programming
(GP) to combine texture and spectral information encoded by image
descriptors. Experiments show that the proposed method yields slightly
better results than the traditional MaxVer approach.

BibTeX:
@inproceedings{Santos2010IGARSS,
  author = {Jefersson Alex dos Santos and Fabio Augusto Faria and Rodrigo Tripodi Calumby and Ricardo da Silva Torres and Rubens A. C. Lamparelli},
  title = {A Genetic Programming approach for coffee crop recognition},
  booktitle = {IEEE International Geoscience and Remote Sensing Symposium (IGARSS)},
  year = {2010},
  pages = {3418--3421},
  doi = {http://doi.org/10.1109/IGARSS.2010.5650273}
}
      


dos Santos, J.A., Penatti, O.A.B. and da S. Torres, R. Evaluating the Potential of Texture and Color Descriptors for Remote Sensing Image Retrieval and Classification 2010
Vol. 25th International Conference on Computer Vision Theory and Applications, pp. 203-208 
inproceedings  
Abstract: 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 may 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 perform experiments
to evaluate the effectiveness of these descriptors, considering image
retrieval and classification tasks. To evaluate descriptors in classification
tasks, we also propose 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.

BibTeX:
@inproceedings{Santos2010VISAPP,
  author = {Jefersson Alex dos Santos and Otávio Augusto Bizetto Penatti and Ricardo da S. Torres},
  title = {Evaluating the Potential of Texture and Color Descriptors for Remote Sensing Image Retrieval and Classification},
  booktitle = {5th International Conference on Computer Vision Theory and Applications},
  year = {2010},
  volume = {2},
  pages = {203--208}
}
      


Almeida, J., Minetto, R., Almeida, T.A., da S. Torres, R. and Leite, N.J. Robust Estimation of Camera Motion using Optical Flow Models 2009 International Symposium on Advances in Visual Computing, pp. 435-446  inproceedings DOI  
Abstract: The estimation of camera motion is one of the most important aspects
for video processing, analysis, indexing, and retrieval. Most of
existing techniques to estimate camera motion are based on optical
flow methods in the uncompressed domain. However, to decode and to
analyze a video sequence is extremely time-consuming. Since video
data are usually available in MPEG-compressed form, it is desirable
to directly process video material without decoding. In this paper,
we present a novel approach for estimating camera motion in MPEG
video sequences. Our technique relies on linear combinations of optical
flow models. The proposed method first creates prototypes of optical
flow, and then performs a linear decomposition on the MPEG motion
vectors, which is used to estimate the camera parameters. Experiments
on synthesized and real-world video clips show that our technique
is more effective than the state-of-the-art approaches for estimating
camera motion in MPEG video sequences.

BibTeX:
@inproceedings{Almeida2009ISVC,
  author = {Jurandy Almeida and Rodrigo Minetto and Tiago A. Almeida and Ricardo da S. Torres and Neucimar J. Leite},
  title = {Robust Estimation of Camera Motion using Optical Flow Models},
  booktitle = {International Symposium on Advances in Visual Computing},
  year = {2009},
  pages = {435--446},
  doi = {http://doi.org/10.1007/978-3-642-10331-5%5C_41}
}
      


Montoya-Zegarra, J.A., Papa, J.P., Leite, N.J., da S. Torres, R. and Falcão, A.X. Novel Approaches for Exclusive and Continuous Fingerprint Classification 2009 3rd Pacific-Rim Symposium on Image and Video Technology, pp. 386-397  inproceedings DOI  
Abstract: This paper proposes novel exclusive and continuous approaches to guide
the search and the retrieval in fingerprint image databases. Both
approaches are useful to perform a coarse level classification of
fingerprint images before fingerprint authentication tasks. Our approaches
are characterized by: (1) texture image descriptors based on pairs
of multi-resolution decomposition methods that encode effectively
global and local fingerprint information, with similarity measures
used for fingerprint matching purposes, and (2) a novel multi-class
object recognition method based on the Optimum Path Forest classifier.
Experiments were carried out on the standard NIST-4 dataset aiming
to study the discriminative and scalability capabilities of our approaches.
The high classification rates allow us demonstrate the feasibility
and validity of our approaches for characterizing fingerprint images
accurately.

BibTeX:
@inproceedings{Montoya-Zegarra2009PSIVT,
  author = {Javier A. Montoya-Zegarra and João P. Papa and Neucimar J. Leite and Ricardo da S. Torres and Alexandre X. Falcão},
  title = {Novel Approaches for Exclusive and Continuous Fingerprint Classification},
  booktitle = {3rd Pacific-Rim Symposium on Image and Video Technology},
  year = {2009},
  pages = {386--397},
  doi = {http://doi.org/10.1007/978-3-540-92957-4%5C_34}
}
      


Murthy, U., Fox, E.A., Chen, Y., Hallerman, E., da Silva Torres, R., Ramos, E.J. and Falcão, T.R.C. Superimposed Image Description and Retrieval for Fish Species Identification 2009 13th European Conference on Digital Libraries, pp. 285-296  inproceedings DOI  
Abstract: Fish species identification is critical to the study of fish ecology
and management of fisheries. Traditionally, dichotomous keys are
used for fish identification. The keys consist of questions about
the observed specimen. Answers to these questions lead to more questions
till the reader identifies the specimen. However, such keys are incapable
of adapting or changing to meet different fish identification approaches,
and often do not focus upon distinguishing characteristics favored
by many field ecologists and more user-friendly field guides. This
makes learning to identify fish difficult for Ichthyology students.
Students usually supplement the use of the key with other methods
such as making personal notes, drawings, annotated fish images, and
more recently, fish information websites, such as Fishbase. Although
these approaches provide useful additional content, it is dispersed
across heterogeneous sources and can be tedious to access. Also,
most of the existing electronic tools have limited support to manage
user created content, especially that related to parts of images
such as markings on drawings and images and associated notes. We
present SuperIDR, a superimposed image description and retrieval
tool, developed to address some of these issues. It allows users
to associate parts of images with text annotations. Later, they can
retrieve images, parts of images, annotations, and image descriptions
through text- and content-based image retrieval. We evaluated SuperIDR
in an undergraduate Ichthyology class as an aid to fish species identification
and found that the use of SuperIDR yielded a higher likelihood of
success in species identification than using traditional methods,
including the dichotomous key, fish web sites, notes, etc.

BibTeX:
@inproceedings{Murthy2009ECDL,
  author = {Uma Murthy and Edward A. Fox and Yinlin Chen and Eric Hallerman and Ricardo da Silva Torres and Evandro J. Ramos and Tiago R. C. Falcão},
  title = {Superimposed Image Description and Retrieval for Fish Species Identification},
  booktitle = {13th European Conference on Digital Libraries},
  year = {2009},
  pages = {285--296},
  doi = {http://doi.org/10.1007/978-3-642-04346-8%5C_28}
}
      


Santos, K.C.L., de Almeida, H.M., Gonçalves, M.A. and da Silva Torres, R. Recuperação de Imagens da Web Utilizando Múltiplas Evidências Textuais e Programação Genética 2009 XXIV Brazilian Symposium on Databases, pp. 91-105  inproceedings URL 
Abstract: This paper describes an image retrieval framework that employs a evolutionary
approach. The proposed framework explores Genetic Programming for
combining multiple textual sources of evidence associated to Web
images. Experiments performed with a collection extracted from the
Web formed by 1,223,829 pages and 254,495 images showed that the
evolutionary framework was able to overcome the Belief Networks (Bayesian
Model) with gains upper to 100% for precision (pn and MAP) and
recall measures.

BibTeX:
@inproceedings{Santos2009SBBD,
  author = {Katia C. L. Santos and Humberto Mossri de Almeida and Marcos André Gonçalves and Ricardo da Silva Torres},
  title = {Recuperação de Imagens da Web Utilizando Múltiplas Evidências Textuais e Programação Genética},
  booktitle = {XXIV Brazilian Symposium on Databases},
  year = {2009},
  pages = {91--105},
  url = {http://www.lbd.dcc.ufmg.br:8080/colecoes/sbbd/2009/007.pdf}
}
      


dos Santos, J.A., Lamparelli, R. and da S. Torres, R. Using Relevance Feedback for Classifying Remote Sensing Images 2009 Brazilian Remote Sensing Symposium, pp. 7909-7916  inproceedings  
Abstract: This paper presents an interactive technique for remote sensing image
classification. In our proposal, users are able to interact with
the classification system, indicating regions which are of interest.
Furthermore, a genetic programming approach is used to learn user
preferences and combine image region descriptors that encode spectral
and texture properties. The approach to classify images can be divided
into four main steps: (i) image partition and region feature extraction,
(ii) identification of the partitions which are of interest, (iii)
image segmentation, and (iv) region vectorization. This work describes
the obtained results from the first two main steps: partition/extraction
of image features and recognition of partitions of interest. So,
in the first step the image are partitioned into tiles. Each tile
is considered as an independent image and this process starts by
the indication of a query image by the user. This query image is
assumed to present the same texture and spectral properties of the
RSI regions which are of interest. A similarity search is performed
and the most similar tiles are returned to the user. The user then
indicates if the returned tiles are relevant or non-relevant. By
using this feedback, the classification system learns the user needs
and tunes itself in order to improve the results in the next iteration.
This process is repeated until the user is satisfied with the result.
Experiments demonstrate that the proposed method is effective and
suitable for image classification tasks.

BibTeX:
@inproceedings{Santos2009SBSR,
  author = {Jefersson A. dos Santos and Rubens Lamparelli and Ricardo da S. Torres},
  title = {Using Relevance Feedback for Classifying Remote Sensing Images},
  booktitle = {Brazilian Remote Sensing Symposium},
  year = {2009},
  pages = {7909--7916}
}
      


Almeida, J.G., Rocha, A., da S. Torres, R. and Goldestein, S. Making Colors Worth more than a Thousand Words 2008 The 23th Annual ACM Symposium on Applied Computing, pp. 1184-1190  inproceedings DOI  
Abstract: 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.

BibTeX:
@inproceedings{Almeida2008ACMSAC,
  author = {Jurandy G. Almeida and Anderson Rocha and Ricardo da S. Torres and Siome Goldestein},
  title = {Making Colors Worth more than a Thousand Words},
  booktitle = {The 23th Annual ACM Symposium on Applied Computing},
  year = {2008},
  pages = {1184--1190},
  doi = {http://doi.org/10.1145/1363686.1363961}
}
      


Escalona-Cuaresma, M.J., Torres-Zenteno, A., Gutierrez, J., Martins, E., da S. Torres, R. and Baranauskas, M.C.C. A Development Process for Web Geographic Information System: A Case of Study 2008
Vol. HCIProceedings of the 10th International Conference on Enterprise Information Systems (ICEIS 2008), pp. 112-117 
inproceedings  
Abstract: This paper introduces a process for developing Web GIS (Geographic
Information Systems) applications. This process integrates the NDT
(Navigational Development Techniques) approach with some of the Organizational
Semiotic models. The use of the proposed development process is illustrated
for a real application: the construction of the WebMaps system. WebMaps
is a Web GIS system whose main goal is to support harvest planning
in Brazil.

BibTeX:
@inproceedings{Escalona-Cuaresma2008ICEIS,
  author = {Maria J. Escalona-Cuaresma and Arturo Torres-Zenteno and Javier Gutierrez and Eliane Martins and Ricardo da S. Torres and Maria Cecília C. Baranauskas},
  title = {A Development Process for Web Geographic Information System: A Case of Study},
  booktitle = {Proceedings of the 10th International Conference on Enterprise Information Systems (ICEIS 2008)},
  publisher = {INSTICC},
  year = {2008},
  volume = {HCI},
  pages = {112--117},
  note = {ISBN: 978-989-8111-40-1}
}
      


Ferreira, C.D., da S. Torres, R., Goncalves, M.A. and Fan, W. Image Retrieval with Relevance Feedback based on Genetic Programming 2008 XXIII Brazilian Symposium on Databases, pp. 120-134  inproceedings  
Abstract: This paper presents a new content-based image retrieval framework
with relevance feedback. This framework employs Genetic Programming
to discover a combination of descriptors that better characterizes
the user perception of image similarity. Several experiments were
conducted to validate the proposed framework. These experiments employed
three different image databases and color, shape, and texture descriptors
to represent the content of database images. The proposed framework
was compared with three other relevance feedback methods regarding
their efficiency and effectiveness in image retrieval tasks. Experiment
results demonstrate the superiority of the proposed method.

BibTeX:
@inproceedings{Ferreira2008SBBD,
  author = {Cristiano D. Ferreira and Ricardo da S. Torres and Marcos A. Goncalves and Weiguo Fan},
  title = {Image Retrieval with Relevance Feedback based on Genetic Programming},
  booktitle = {XXIII Brazilian Symposium on Databases},
  year = {2008},
  pages = {120--134}
}
      


Montoya-Zegarra, J.A., Beek-Pepper, J.C., Leite, N.J., da S. Torres, R. and Falcão, A.X. Combining Global with Local Texture Information for Image Retrieval Applications 2008 IEEE International Symposium on Multimedia, pp. 148-153  inproceedings DOI  
Abstract: This paper proposes a new texture descriptor to guide the search and
retrieval in image databases. It extracts rich information from global
and local primitives of textured images. At a higher level, the global
macro-features in textured images are characterized by exploiting
the multiresolution properties of the Steerable Pyramid Decomposition.
By doing this, the global texture configurations are highlighted.
At a finer level, the local arrangements of texture micro-patterns
are encoded by the Local Binary Pattern operator.Experiments were
carried out on the standard Vistex dataset aiming to compare our
descriptors against popular texture extraction methods with regard
to their retrieval accuracies. The comparative evaluations allowed
us to show the superior descriptive properties of our feature representation
methods.

BibTeX:
@inproceedings{Montoya-Zegarra2008ISM,
  author = {Javier A. Montoya-Zegarra and Jan C. Beek-Pepper and Neucimar J. Leite and Ricardo da S. Torres and Alexandre X. Falcão},
  title = {Combining Global with Local Texture Information for Image Retrieval Applications},
  booktitle = {IEEE International Symposium on Multimedia},
  year = {2008},
  pages = {148--153},
  doi = {http://doi.org/10.1109/ISM.2008.113}
}
      


Moreno, M.A.M. and da S. Torres, R. Recuperação de Imagem Utilizando Descritores Baseados em Esqueleto 2008 IV Workshop de Visão Computacional  inproceedings  
Abstract: Este artigo apresenta duas técnicas de caracteriza¸ cão de formas
baseada em esqueletos. Experimentos ocnduzidos mostram um aumento
da eficácia quando as técnicas propostas são combinadas com
descritores de forma tradicionais.

BibTeX:
@inproceedings{Moreno2008WVC,
  author = {Marcio A. M. Moreno and Ricardo da S. Torres},
  title = {Recuperação de Imagem Utilizando Descritores Baseados em Esqueleto},
  booktitle = {IV Workshop de Visão Computacional},
  year = {2008}
}
      


Pedronette, D.C.G. and da S. Torres, R. Uma Plataforma de Serviços de Recomendação para Bibliotecas Digitais 2008 XIII Brazilian Symposium on Databases, pp. 253-267  inproceedings  
Abstract: This paper presents a platform for recommendation services, called
RecS-DL, to support the use of recommendation tools by digital libray
applications. The proposed RecS-DL platform is independent of application
domain, technology, and recommendation techniques. The recommendation
services offered by the platform can be easily incorporated into
digital libraries systems. Furthermore, new recommendation engines
can also be plugged into the platform in a dynamic way. We present
the results obtained from experiments conducted with real digital
libraries and from evaluations made by potential users. Experimental
results show that the platform facilitates the interoperability of
recommendation tools in digital libraries systems.

BibTeX:
@inproceedings{Pedronette2008SBBD,
  author = {Daniel C. G. Pedronette and Ricardo da S. Torres},
  title = {Uma Plataforma de Serviços de Recomendação para Bibliotecas Digitais},
  booktitle = {XIII Brazilian Symposium on Databases},
  year = {2008},
  pages = {253--267}
}
      


Penatti, O.B. and da S. Torres, R. Color Descriptors for Web Image Retrieval: a Comparative Study 2008 XXI Brazilian Symposium on Computer Graphics and Image Processing, pp. 163-170  inproceedings DOI  
Abstract: This paper presents a comparative study of color descriptors for content-based
image retrieval on the Web. Several image descriptors were compared
theoretically and the most relevant ones were implemented and tested
in two different databases. The main goal was to find out the best
descriptors for Web image retrieval. Descriptors are compared according
to the extraction and distance functions complexities, the compactness
of feature vectors, and the ability to retrieve relevant images.

BibTeX:
@inproceedings{Penatti2008SIBGRAPI,
  author = {Otávio B. Penatti and Ricardo da S. Torres},
  title = {Color Descriptors for Web Image Retrieval: a Comparative Study},
  booktitle = {XXI Brazilian Symposium on Computer Graphics and Image Processing},
  year = {2008},
  pages = {163--170},
  doi = {http://doi.org/10.1109/SIBGRAPI.2008.20}
}
      


Rocha, A., Almeida, J.G., Nascimento, M., da S. Torres, R. and Goldenstein, S. Efficient and Flexible Cluster-and-Search for CBIR 2008 Advanced Concepts for Intelligent Vision Systems, pp. 77-88  inproceedings DOI  
Abstract: Content-Based Image Retrieval is a challenging problem both in terms
of effectiveness and efficiency. In this paper, we present a flexible
cluster-and-search approach that is able to reuse any previously
proposed image descriptor as long as a suitable similarity function
is provided. In the clustering step, the image data set is clustered
using a hybrid divisive-agglomerative hierarchical clustering technique.
The obtained clusters are organized in a tree that can be traversed
efficiently using the similarity function associated with the chosen
image descriptors. Our experiments have shown that we can improve
search-time performance by a factor of 10 or more, at the cost of
small loss in effectiveness (typically less than 15%) when compared
to the state-of-the-art solutions.

BibTeX:
@inproceedings{Rocha2008ACIVS,
  author = {Anderson Rocha and Jurandy G. Almeida and Mario Nascimento and Ricardo da S. Torres and Siome Goldenstein},
  title = {Efficient and Flexible Cluster-and-Search for CBIR},
  booktitle = {Advanced Concepts for Intelligent Vision Systems},
  year = {2008},
  pages = {77--88},
  doi = {http://doi.org/10.1007/978-3-540-88458-3%5C_8}
}
      


dos Santos, J.A., Ferreira, C.D. and da S. Torres, R. A Genetic Programming Approach for Relevance Feedback in Region-based Image Retrieval Systems 2008 XXI Brazilian Symposium on Computer Graphics and Image Processing, pp. 155-162  inproceedings DOI  
Abstract: 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.

BibTeX:
@inproceedings{Santos2008SIBGRAPI,
  author = {Jefersson A. dos Santos and Cristiano D. Ferreira and Ricardo da S. Torres},
  title = {A Genetic Programming Approach for Relevance Feedback in Region-based Image Retrieval Systems},
  booktitle = {XXI Brazilian Symposium on Computer Graphics and Image Processing},
  year = {2008},
  pages = {155--162},
  doi = {http://doi.org/10.1109/SIBGRAPI.2008.15}
}
      


da S. Torres, R., Zegarra, J.A.M., Santos, J.A., Ferreira, C.D., Penatti, O.A.B., Andaló, F.A. and Almeida, J.G. Recuperação de Imagens: Desafios e Novos Rumos 2008 XXXV Seminário Integrado de Software e Hardware (SEMISH), pp. 223-237  inproceedings  
Abstract: Huge image collections have been created, managed and stored into
image databases. Given the large size of these collections it is
essential to provide efficient and effective mechanisms to retrieve
images. This is the objective of the so-called content-based image
retrieval -- CBIR -- systems. Traditionally, these systems are based
on objective criteria to represent and compare images. However, users
of CBIR systems tend to use subjective elements to compare images.
The use of these elements have improved the effectiveness of content-
based image retrieval systems. This paper discusses approaches that
incorpo- rate semantic information into content-based image retrieval
process, highlighting some new challenges on this area.

BibTeX:
@inproceedings{Torres2008SEMISH,
  author = {Ricardo da S. Torres and Javier A. M. Zegarra and Jefersson A. Santos and Cristiano D. Ferreira and Otávio A. B. Penatti and Fernanda A. Andaló and Jurandy G. Almeida},
  title = {Recuperação de Imagens: Desafios e Novos Rumos},
  booktitle = {XXXV Seminário Integrado de Software e Hardware (SEMISH)},
  year = {2008},
  pages = {223--237}
}
      


Andaló, F.A., Miranda, P.A.V., da S. Torres, R. and Falcão, A.X. Detecting Contour Saliences Using Tensor Scale 2007 IEEE International Conference on Image Processing, pp. VI349-VI352  inproceedings DOI  
Abstract: Tensor scale is a morphometric parameter that unifies the representation
of local structure thickness, orientation, and anisotropy, which
can be used in several image processing tasks. This paper introduces
a new application for tensor scale, which is the detection of saliences
on a given contour, based on the tensor scale orientations computed
for the entire object and mapped to its contour. For validation purposes,
we present a shape descriptor that uses the detected contour saliences.
Experimental results are provided, comparing the proposed method
with our previous contour salience descriptor (CS). We show that
the proposed method can be not only faster and more robust in the
detection of salience points than the CS method, but also more effective
as a shape descriptor.

BibTeX:
@inproceedings{Andalo2007ICIP,
  author = {Fernanda A. Andaló and Paulo A. V. Miranda and Ricardo da S. Torres and Alexandre X. Falcão},
  title = {Detecting Contour Saliences Using Tensor Scale},
  booktitle = {IEEE International Conference on Image Processing},
  year = {2007},
  pages = {VI349--VI352},
  doi = {http://doi.org/10.1109/ICIP.2007.4379593}
}
      


Andaló, F.A., Miranda, P.A.V., da S. Torres, R. and Falcão, A.X. A New Shape Descriptor based on Tensor Scale 2007 8th International Symposium on Mathematical Morphology, pp. 141-152  inproceedings  
Abstract: Tensor scale is a morphometric parameter that unifies the representation
of local structure thickness, orientation, and anisotropy, which
can be used in several computer vision and image processing tasks.
In this paper, we exploit this concept for binary images and propose
a shape descriptor that encodes region and contour properties in
a very efficient way. Experimental results are provided, showing
the effectiveness of the proposed descriptor, when compared to other
relevant shape descriptors, with regard to their use in content-based
image retrieval systems.

BibTeX:
@inproceedings{Andalo2007ISMM,
  author = {Fernanda A. Andaló and Paulo A. V. Miranda and Ricardo da S. Torres and Alexandre X. Falcão},
  title = {A New Shape Descriptor based on Tensor Scale},
  booktitle = {8th International Symposium on Mathematical Morphology},
  year = {2007},
  pages = {141--152}
}
      


Mariote, L., Medeiros, C.B. and da S. Torres, R. Diagnosing Similarity of Oscillation Trends in Time Series 2007 International Workshop on Spatial and Spatio-Temporal Data Mining, 7th IEEE International Conference on Data Mining (ICDM 2007), pp. 643-648  inproceedings DOI  
Abstract: Sensor networks have increased the amount and variety of temporal
data available, requiring the definition of new techniques for data
mining. Related research typically addresses the problems of indexing,
clustering, classification, summarization, and anomaly detection.
They present many ways for describing and comparing time series,
but they focus on their values. This paper concentrates on a new
aspect that of describing oscillation patterns. It presents a technique
for time series similarity search, based on multiple temporal scales,
defining a descriptor that uses the angular coefficients from a linear
segmentation of the curve that represents the evolution of the analyzed
series. Preliminary experiments with real datasets showed that our
approach correctly characterizes the oscillation of time series.

BibTeX:
@inproceedings{Mariote2007SSTDM,
  author = {Leonardo Mariote and Claudia B. Medeiros and R. da S. Torres},
  title = {Diagnosing Similarity of Oscillation Trends in Time Series},
  booktitle = {International Workshop on Spatial and Spatio-Temporal Data Mining, 7th IEEE International Conference on Data Mining (ICDM 2007)},
  year = {2007},
  pages = {643--648},
  doi = {http://doi.org/10.1109/ICDMW.2007.28}
}
      


Montoya-Zegarra, J.A., Papa, J.P., Leite, N.J., da S. Torres, R. and Falcão, A.X. Rotation-invariant Texture Recognition 2007 International Symposium on Visual Computing, pp. 193-204  inproceedings DOI  
Abstract: This paper proposes a new texture classification system, which is
distinguished by: (1) a new rotation-invariant image descriptor based
on Steerable Pyramid Decomposition, and (2) by a novel multi-class
recognition method based on Optimum Path Forest. By combining the
discriminating power of our image descriptor and classifier, our
system uses small size feature vectors to characterize texture images
without compromising overall classification rates. State-of-the-art
recognition results are further presented on the Brodatz dataset.
High classification rates demonstrate the superiority of the proposed
method.

BibTeX:
@inproceedings{Montoya-Zegarra2007ISVC,
  author = {Javier A. Montoya-Zegarra and João P. Papa and Neucimar J. Leite and Ricardo da S. Torres and A. X. Falcão},
  title = {Rotation-invariant Texture Recognition},
  booktitle = {International Symposium on Visual Computing},
  year = {2007},
  pages = {193--204},
  doi = {http://doi.org/10.1007/978-3-540-76856-2%5C_19}
}
      


Montoya-Zegarra, J.A., Leite, N.J. and da S. Torres, R. Rotation-Invariant and Scale-Invariant Steerable Pyramid Decomposition for Texture Image Retrieval 2007 XX Brazilian Symposium on Computer Graphics and Image Processing, pp. 121-128  inproceedings DOI  
Abstract: This paper proposes a new rotation-invariant and scale-invariant representation
for texture image retrieval based on steerable pyramid decomposition.
By calculating the mean and standard deviation of decomposed image
subbands, the texture feature vectors are extracted. To obtain rotation
or scale invariance, the feature elements are aligned by considering
either the dominant orientation or dominant scale of the input textures.
Experiments were conducted on the Brodatz database aiming to compare
our approach to the conventional steerable pyramid decomposition,
and a proposal for texture characterization based on Gabor wavelets
with regard to their retrieval effectiveness. Results demonstrate
the superiority of the proposed method in rotated and scaled image
datasets.

BibTeX:
@inproceedings{Montoya-Zegarra2007SIBGRAPI,
  author = {Javier A. Montoya-Zegarra and Neucimar J. Leite and Ricardo da S. Torres},
  title = {Rotation-Invariant and Scale-Invariant Steerable Pyramid Decomposition for Texture Image Retrieval},
  booktitle = {XX Brazilian Symposium on Computer Graphics and Image Processing},
  year = {2007},
  pages = {121--128},
  doi = {http://doi.org/10.1109/SIBGRAPI.2007.42}
}
      


Murthy, U., Gorton, D., da S. Torres, R., calves , M.A.G., Fox, E.A. and Delcambre, L. Extending the 5S Digital Library (DL) Framework: From a Minimal DL towards a DL Reference Model 2007 1st Workshop on Digital Library Foundations, ACM IEEE Joint Conference on Digital Libraries  inproceedings  
Abstract: In this paper, we describe ongoing research in three DL projects that
build upon a common foundation - the 5S DL framework. In each project,
we extend the 5S framework to provide specifications for a particular
type of DL service and/or system - finally, moving towards a DL reference
model. In the first project, we are working on formalizing content-based
image retrieval services in a DL. In the second project, we are developing
specifications for a superimposed information-supported DL (combining
annotation, hypertext, and knowledge management technologies). In
the third effort, we have used the 5S framework to generate a practical
DL system based on the DSpace software.

BibTeX:
@inproceedings{Murthy2007WDLF-JCDL,
  author = {Uma Murthy and Douglas Gorton and Ricardo da S. Torres and Marcos A. Gon¸ calves and Edward A. Fox and Lois Delcambre},
  title = {Extending the 5S Digital Library (DL) Framework: From a Minimal DL towards a DL Reference Model},
  booktitle = {1st Workshop on Digital Library Foundations, ACM IEEE Joint Conference on Digital Libraries},
  year = {2007}
}
      


Penatti, O.B. and da S. Torres, R. Descritor de Relacionamento Espacial baseado em Parti¸ cões 2007 XXVI Concurso de Trabalhos de Inicia¸ cão Científica, XXVII Congresso da Sociedade Brasileira de Computa¸ cão   inproceedings  
Abstract: Spatial relationships can be fundamental for image recognition and
retrieval, being useful for geographic and medical applications,
for instance. This paper presents a new spatial relationship descriptor
for content-based image retrieval. The new descriptor presented is
based on partitioning the space of analysis into quadrants and on
counting the number of object points in each partition. Experiments
demonstrated that the proposed descriptor is more effective than
the other descriptors studied in this work.

BibTeX:
@inproceedings{Penatti2007CTIC-SBC,
  author = {Otávio B. Penatti and Ricardo da S. Torres},
  title = {Descritor de Relacionamento Espacial baseado em Parti¸ cões},
  booktitle = {XXVI Concurso de Trabalhos de Inicia¸ cão Científica, XXVII Congresso da Sociedade Brasileira de Computa¸ cão },
  year = {2007}
}
      


Shen, R., Vemuri, N.S., Fan, W., da S. Torres, R. and Fox, E.A. Exploring Digital Libraries: Integrating Browsing, Searching, and Visualization 2006 Proceedings of the 6th ACM/IEEE Joint Conference on Digital libraries, pp. 1-10  inproceedings DOI  
Abstract: Exploring services for digital libraries (DLs) include two major paradigms,
browsing and searching, as well as other services such as clustering
and visualization. In this paper, we formalize and generalize DL
exploring services within a DL theory. We develop theorems to indicate
that browsing and searching can be converted or mapped to each other
under certain conditions. The theorems guide the design and implementation
of exploring services for an integrated archaeological DL, ETANA-DL.
Its integrated browsing and searching can support users in moving
seamlessly between these operations, minimizing context switching,
and keeping users focused. It also integrates browsing and searching
into a single visual interface for DL exploration. A user study to
evaluate ETANA-DL's exploring services helped validate our hypotheses.

BibTeX:
@inproceedings{Shen2006JCDL,
  author = {Rao Shen and Naga S. Vemuri and Weiguo Fan and Ricardo da S. Torres and Edward A. Fox},
  title = {Exploring Digital Libraries: Integrating Browsing, Searching, and Visualization},
  booktitle = {Proceedings of the 6th ACM/IEEE Joint Conference on Digital libraries},
  year = {2006},
  pages = {1--10},
  doi = {http://doi.org/10.1145/1141753.1141755}
}
      


Carvalho, A.C.P.L.F., Brayner, A., Loureiro, A., Furtado, A.L., v. Staa, A., Lucena, C.J.P., C. S., S., Medeiros, C.M.B., Lucchesi, C.L., Silva, E.S., Wagner, F.R., Simon, I., Wainer, J., Maldonado, J.C., Oliveira, J.P.M., Ribeiro, L., Velho, L., calves , M.A.G., Baranauskas, M.C.C., Mattoso, M., Ziviani, N., Navaux, P.O.A., da S. Torres, R., Almeida, V.A.F., Jr., W.M. and Kohayakawa, Y. Grandes Desafios da Pesquisa em Computa\ {a}o no Brasil -- 2006 - 2016 2006 Seminário Grandes Desafios da Sociedade Brasileira de Computação  inproceedings  
BibTeX:
@inproceedings{Torres2006GrandesDesafios,
  author = {A. C. P. L. F. Carvalho and A. Brayner and A. Loureiro and A. L. Furtado and A. v. Staa and C. J. P. Lucena and C. S. S. and C. M. B. Medeiros and C. L. Lucchesi and E. S. Silva and F. R. Wagner and I. Simon and J. Wainer and J. C. Maldonado and J. P. M. Oliveira and L. Ribeiro and L. Velho and M. A. Gon¸ calves and M. C. C. Baranauskas and M. Mattoso and N. Ziviani and P. O. A. Navaux and R. da S. Torres and V. A. F. Almeida and W. Meira Jr. and Y. Kohayakawa},
  title = {Grandes Desafios da Pesquisa em Computa\ {a}o no Brasil -- 2006 - 2016},
  booktitle = {Seminário Grandes Desafios da Sociedade Brasileira de Computação},
  year = {2006}
}
      


da. da S. Torres, R. Sistemas de Informação para o Gerenciamento de Imagens: Aplicações e Desafios de Pesquisa 2006 Grandes Desafios da Sociedade Brasileira de Computação  inproceedings  
BibTeX:
@inproceedings{Torres2006sbc,
  author = {Ricardo da. da S. Torres},
  title = {Sistemas de Informação para o Gerenciamento de Imagens: Aplicações e Desafios de Pesquisa},
  booktitle = {Grandes Desafios da Sociedade Brasileira de Computação},
  year = {2006}
}
      


Torres-Zenteno, A.H., Martins, E., da S. Torres, R. and Escalona-Cuaresma, M.J. Teste de Desempenho em Aplicações SIG Web 2006 IX Conferencia Iberoamericana de Software Engineering (CIbSE 2006), pp. 449-462  inproceedings  
Abstract: Este artigo propõe um modelo de processo de teste de desempenho
para aplica¸ cões SIG Web. O modelo considera os casos de uso
mais críticos ou de maior risco quanto ao desempenho de um sistema
para a cria¸ cão de cenários de testes. Além disso, prevê
a utiliza¸ cão de ferramentas livres para automatiza¸ cão de
etapas do processo de avalia¸ cão. O modelo foi aplicado ao projeto
WebMaps, que é uma aplica¸ cão SIG Web cuja finalidade é auxiliar
seus usuários no planejamento agrícola a partir de regiões
de interesse. Os resultados preliminares obtidos indicam que os testes
foram úteis na identifica¸ cão de problemas da arquitetura preliminar
do sistema.

BibTeX:
@inproceedings{Torres-Zenteno2006IDEAS,
  author = {Arturo Henry Torres-Zenteno and Eliane Martins and Ricardo da S. Torres and María José Escalona-Cuaresma},
  title = {Teste de Desempenho em Aplicações SIG Web},
  booktitle = {IX Conferencia Iberoamericana de Software Engineering (CIbSE 2006)},
  year = {2006},
  pages = {449-462}
}
      


Freitas, R.B. and da S. Torres, R. OntoSAIA: Um ambiente Baseado em Ontologias para Recuperação e Anotação Semi-Automática de Imagens 2005 First Workshop on Digital Libraries (Primeiro Workshop de Bibliotecas Digitais), XX Brazilian Symposium on Databases, pp. 60-69  inproceedings  
Abstract: This article proposes the use of image content, keywords and ontologies
to improve the image annotation and retrieval processes through the
enhancement of the user's knowledge of an image database. It proposes
an architecture of a flexible system capable of dealing with multiple
ontologies and multiple image content descriptors to help these tasks.
The validation of the idea is being done through the implementation,
in Java, of the software OntoSAIA.

BibTeX:
@inproceedings{Freitas2005WDL-SBBD,
  author = {Ricardo B. Freitas and Ricardo da S. Torres},
  title = {OntoSAIA: Um ambiente Baseado em Ontologias para Recuperação e Anotação Semi-Automática de Imagens},
  booktitle = {First Workshop on Digital Libraries (Primeiro Workshop de Bibliotecas Digitais), XX Brazilian Symposium on Databases},
  year = {2005},
  pages = {60--69}
}
      


Miranda, P.A.V., da S. Torres, R. and Falcão, A.X. TSD: A Shape Descriptor Based on a Distribution of Tensor Scale Local Orientation 2005 18th Brazilian Symposium on Computer Graphics and Image, pp. 139-146  inproceedings DOI  
Abstract: We present tensor scale descriptor (TSD) -- a shape descriptor for
content-based image retrieval, registration, and analysis. TSD exploits
the notion of local structure thickness, orientation, and anisotropy
as represented by the largest ellipse centered at each image pixel
and within the same homogeneous region. The proposed method uses
the normalized histogram of the local orientation (the angle of the
ellipse) at regions of high anisotropy and thickness within a certain
interval. It is shown that TSD is invariant to rotation and to some
reasonable level of scale changes. Experimental results with a fish
database are presented to illustrate and validate the method.

BibTeX:
@inproceedings{Miranda2005SIBGRAPI,
  author = {Paulo A. V. Miranda and Ricardo da S. Torres and Alexandre X. Falcão},
  title = {TSD: A Shape Descriptor Based on a Distribution of Tensor Scale Local Orientation},
  booktitle = {18th Brazilian Symposium on Computer Graphics and Image},
  year = {2005},
  pages = {139--146},
  doi = {http://doi.org/10.1109/SIBGRAPI.2005.51}
}
      


da S. Torres, R., Silva, C.G., Medeiros, C.B. and da Rocha, H.V. Visual structures for image browsing 2003 Proceedings of the 12th International Conference on Information and Knowledge Management, pp. 49-55  inproceedings DOI  
Abstract: Content-Based Image Retrieval (CBIR) presents several challenges and
has been subject to extensive research from many domains, such as
image processing or database systems. Database researchers are concerned
with indexing and querying, whereas image processing experts worry
about extracting appropriate image descriptors. Comparatively little
work has been done on designing user interfaces for CBIR systems.
This, in turn, has a profound effect on these systems since the concept
of image similarity is strongly influenced by user perception. This
paper describes an initial effort to fill this gap, combining recent
research in CBIR and Information Visualization, studied from a Human-Computer
Interface perspective. It presents two visualization techniques based
on Spiral and Concentric Rings implemented in a CBIR system to explore
query results. The approach is centered on keeping user focus on
both the query image, and the most similar retrieved images. Experiments
conducted so far suggest that the proposed visualization strategies
improves system usability.

BibTeX:
@inproceedings{Torres2003CIKM,
  author = {Ricardo da S. Torres and Celmar G. Silva and Claudia Bauzer Medeiros and Heloisa Vieira da Rocha},
  title = {Visual structures for image browsing},
  booktitle = {Proceedings of the 12th International Conference on Information and Knowledge Management},
  year = {2003},
  pages = {49--55},
  doi = {http://doi.org/10.1145/956863.956874}
}
      


da S. Torres, R., Picado, E.M., Falcão, A.X. and da Fontoura Costa, L. Effective Image Retrieval by Shape Saliences 2003 16th Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI 2003), pp. 167-174  inproceedings DOI  
Abstract: Content-based image retrieval (CBIR) systems have been developed aiming
at enabling users to search and retrieve images based on their properties
such as shape, color and texture. We are concerned with shape-based
image retrieval. Here, we discuss a recently proposed shape descriptor,
called contour saliences, defined as the influence areas of its higher
curvature points. We introduce a robust approach to estimate contour
saliences by exploiting the relation between a contour and its skeleton,
modifies the original definition to include the location and the
value of saliences along the contour, and proposes a new metric to
compare contour saliences. We also evaluate the effectiveness of
the proposed descriptor with respect to Fourier descriptors, curvature
scale space and moment invariants.

BibTeX:
@inproceedings{Torres2003SIBGRAPI,
  author = {Ricardo da S. Torres and Eduardo M. Picado and Alexandre X. Falcão and Luciano da Fontoura Costa},
  title = {Effective Image Retrieval by Shape Saliences},
  booktitle = {16th Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI 2003)},
  year = {2003},
  pages = {167--174},
  doi = {http://doi.org/10.1109/SIBGRA.2003.1241005}
}
      


da S. Torres, R., Falcão, A.X. and da F. Costa, L. Shape Description by Image Foresting Transform 2002
Vol. 2Proceedings of the 14th International Conference on Digital Signal Processing, pp. 1089-1092 
inproceedings DOI  
Abstract: The image foresting transform (IFT) is a unified and effective graph-based
approach to the design of image-processing operations, often with
considerable efficiency gains over published algorithms. This paper
extends the applications of the Euclidean IFT to two recently proposed
shape descriptors: saliences and multiscale fractal dimension. It
explains how to obtain the salience information and the multiscale
fractal dimension of contours and skeletons and presents their comparison
in terms of robustness and separability.

BibTeX:
@inproceedings{Torres2002DSP,
  author = {Ricardo da S. Torres and Alexandre X. Falcão and Luciano da F. Costa},
  title = {Shape Description by Image Foresting Transform},
  booktitle = {Proceedings of the 14th International Conference on Digital Signal Processing},
  year = {2002},
  volume = {2},
  pages = {1089--1092},
  doi = {http://doi.org/10.1109/ICDSP.2002.1028280}
}