FUSION DE DATOS CONSIDERANDO LAS DISIMILITUDES Y OTROS TIPOS DE RELACIONES ENTRE LOS MISMOS Y APLICACION A INTELIGENCIA ARTIFICIAL

PID2019-108392GB-I00

Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de Generación de Conocimiento y Fortalecimiento Científico y Tecnológico del Sistema de I+D+i
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos I+D
Año convocatoria 2019
Unidad de gestión Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020
Centro beneficiario UNIVERSIDAD PUBLICA DE NAVARRA
Identificador persistente http://dx.doi.org/10.13039/501100011033

Publicaciones

Found(s) 8 result(s)
Found(s) 1 page(s)

A survey on matching strategies for boundary image comparison and evaluation

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
  • Lopez Molina, Carlos
  • Marco-Detchart, Cedric|||0000-0002-4310-9060
  • Bustince, H.
  • De Baets, B.
[EN] Most of the strategies for boundary image evaluation involve the comparison of computer-generated images with ground truth solutions. While this can be done in different manners, recent years have seen a dominance of techniques based on the use of confusion matrices. That is, techniques that, at the evaluation stage, interpret boundary detection as a classification problem. These techniques require a correspondence between the boundary pixels in the candidate image and those in the ground truth; that correspondence is further used to create the confusion matrix, from which evaluation statistics can be computed. The correspondence between boundary images faces different challenges, mainly related to the matching of potentially displaced boundaries. Interestingly, boundary image comparison relates to many other fields of study in literature, from object tracking to biometrical identification. In this work, we survey all existing strategies for boundary matching, we propose a taxonomy to embrace them all, and perform a usability-driven quantitative analysis of their behaviour., The authors gratefully acknowledge the financial support of the Spanish Ministry of Science (PID2019-108392GB-I00, AEI/10.13039/50110 0 011033), ass well as that of the Research Foundation Flanders (FWO project 3G.0838.12.N)




Neuro-inspired edge feature fusion using Choquet integrals

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
  • Marco-Detchart, Cédric|||0000-0002-4310-9060
  • Lucca, Giancarlo
  • Lopez-Molina, Carlos
  • De Miguel, Laura
  • Pereira Dimuro, Graçaliz
  • Bustince, Humberto
[EN] It is known that the human visual system performs a hierarchical information process in
which early vision cues (or primitives) are fused in the visual cortex to compose complex
shapes and descriptors. While different aspects of the process have been extensively stud-
ied, such as lens adaptation or feature detection, some other aspects, such as feature fusion,
have been mostly left aside. In this work, we elaborate on the fusion of early vision prim-
itives using generalizations of the Choquet integral, and novel aggregation operators that
have been extensively studied in recent years. We propose to use generalizations of the
Choquet integral to sensibly fuse elementary edge cues, in an attempt to model the behaviour of neurons in the early visual cortex. Our proposal leads to a fully-framed edge detection algorithm whose performance is put to the test in state-of-the-art edge detection
datasets., The authors gratefully acknowledge the financial support of the Spanish Ministry of Science and Technology (project
PID2019-108392GB-I00 (AEI/10.13039/501100011033), the Research Services of Universidad Pública de Navarra, CNPq
(307781/2016-0, 301618/2019-4), FAPERGS (19/2551-0001660) and PNPD/CAPES (464880/2019-00).




On the normalization of interval data

Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
  • Santiago, Regivan
  • Bergamaschi, Flaulles
  • 0000-0002-1279-6195
  • 0000-0001-6986-9888
  • 0000-0002-7066-7156
  • 0000-0002-1427-9909
The impreciseness of numeric input data can be expressed by intervals. On the other hand, the normalization of numeric data is a usual process in many applications. How do we match the normalization with impreciseness on numeric data? A straightforward answer is that it is enough to apply a correct interval arithmetic, since the normalized exact value will be enclosed in the resulting 'normalized' interval. This paper shows that this approach is not enough since the resulting 'normalized' interval can be even wider than the input intervals. So, we propose a pair of axioms that must be satisfied by an interval arithmetic in order to be applied in the normalization of intervals. We show how some known interval arithmetics behave with respect to these axioms. The paper ends with a discussion about the current paradigm of interval computations., This study was funded by National Council for Scientific and Technological Development (CNPq) within the project 312053/2018-5, by Coordination for the Improvement of Higher Education Personnel (CAPES) within the project Capes-Print 88887.363001/2019-00. Ministerio de Ciencia y innovación within the project PID2019-108392GB-I00 (AEI/10.13039/501100011033).




Constructing interval-valued fuzzy material implication functions derived from general interval-valued grouping functions

Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
  • Pereira Dimuro, Graçaliz
  • Santos, Helida
  • 0000-0002-7066-7156
  • Wieczynski, Jonata
  • Pinheiro, Jocivania
  • 0000-0002-6757-7934
  • 0000-0002-1279-6195
Grouping functions and their dual counterpart,
overlap functions, have drawn the attention of many authors,
mainly because they constitute a richer class of operators compared to other types of aggregation functions. Grouping functions
are a useful theoretical tool to be applied in various problems, like
decision making based on fuzzy preference relations. In pairwise
comparisons, for instance, those functions allow one to convey
the measure of the amount of evidence in favor of either of two
given alternatives. Recently, some generalizations of grouping
functions were proposed, such as (i) the n-dimensional grouping
functions and the more flexible general grouping functions, which
allowed their application in n-dimensional problems, and (ii)
n-dimensional and general interval-valued grouping functions,
in order to handle uncertainty on the definition of the membership functions in real-life problems. Taking into account
the importance of interval-valued fuzzy implication functions in
several application problems under uncertainty, such as fuzzy
inference mechanisms, this paper aims at introducing a new
class of interval-valued fuzzy material implication functions. We
study their properties, characterizations, construction methods
and provide examples., upported by CNPq (301618/2019-4, 311429/2020-3), FAPERGS (19/2551-0001660-3), UFERSA, the Spanish Ministry of Science and Technology (TIN2016-77356-P, PID2019-108392GB I00 (MCIN/AEI/10.13039/501100011033)) and Navarra de Servicios y Tecnologías, S.A. (NASERTIC).




Multi-temporal data augmentation for high frequency satellite imagery: a case study in Sentinel-1 and Sentinel-2 building and road segmentation

Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
  • Ayala Lauroba, Christian
  • Aranda Magallón, Coral
  • 0000-0003-2865-6549
Semantic segmentation of remote sensing images has many practical applications such as urban planning or disaster assessment.
Deep learning-based approaches have shown their usefulness in automatically segmenting large remote sensing images, helping
to automatize these tasks. However, deep learning models require large amounts of labeled data to generalize well to unseen
scenarios. The generation of global-scale remote sensing datasets with high intraclass variability presents a major challenge. For
this reason, data augmentation techniques have been widely applied to artificially increase the size of the datasets. Among them,
photometric data augmentation techniques such as random brightness, contrast, saturation, and hue have been traditionally applied
aiming at improving the generalization against color spectrum variations, but they can have a negative effect on the model due
to their synthetic nature. To solve this issue, sensors with high revisit times such as Sentinel-1 and Sentinel-2 can be exploited
to realistically augment the dataset. Accordingly, this paper sets out a novel realistic multi-temporal color data augmentation
technique. The proposed methodology has been evaluated in the building and road semantic segmentation tasks, considering a
dataset composed of 38 Spanish cities. As a result, the experimental study shows the usefulness of the proposed multi-temporal
data augmentation technique, which can be further improved with traditional photometric transformations., Christian Ayala was partially supported by the Government
of Navarra under the industrial Ph.D. program 2020 reference
0011-1408-2020-000008. Mikel Galar was partially supported
by Tracasa Instrumental S.L. under the project OTRI 2020-
901-156, and by the Spanish MICIN (PID2019-108392GB-I00
/ AEI / 10.13039/501100011033).




Measures of embedding for interval-valued fuzzy sets

Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
  • Bouchet, Agustina
  • 0000-0002-2949-7909
  • 0000-0002-0221-4171
  • 0000-0002-1279-6195
  • Montes, Susana
  • Díaz, Irene
Interval-valued fuzzy sets are a generalization of classical fuzzy sets where the membership values are intervals. The epistemic interpretation of interval-valued fuzzy sets assumes that there is one real-valued membership degree of an element within the membership interval of possible membership degrees. Considering this epistemic interpretation, we propose a new measure, called IV-embedding, to compare the precision of two interval-valued fuzzy sets. An axiomatic definition for this concept as well as a construction method are provided. The construction method is based on aggregation operators and the concept of interval embedding, which is also introduced and deeply studied., Authors would like to thank for the support of the Spanish Ministry of Science and Technology project TIN2017-87600-P (I. Díaz and A. Bouchet), the Spanish Ministry of Science and Technology projects PGC2018-098623-B-I00 (S. Montes) and PID2019-108392GBI00(AEI/10.13039/501100011033) (H. Bustince, G. Ochoa, M. Sesma-Sara).




On the notion of fuzzy dispersion measure and its application to triangular fuzzy numbers

Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
  • Roldán López de Hierro, Antonio Francisco
  • 0000-0002-1279-6195
  • Rueda, María del Mar
  • Roldán, Concepción
  • 0000-0002-7665-2801
  • Guerra, Carlos
In this paper, based on the analysis of the most widely used dispersion measure in the real context (namely, the variance), we introduce the notion of fuzzy dispersion measure associated to a finite set of data given by fuzzy numbers. This measure is implemented as a fuzzy number, so there is no loss of information caused by any defuzzification. The proposed concept satisfies the usual properties in a genuinely fuzzy sense and it avoids limitations in terms of its geometric shape or its analytical properties: under this conception, it could have a piece of its support in the negative part of the real line. This novel notion can be interpreted as a way of fusing the information included in a fuzzy data set in order to make a decision based on its dispersion. To illustrate the main characteristics of this approach, we present an example of a fuzzy dispersion measure that allows to conclude that this new way to deal this problem is coherent, at least, from the point of view of human intuition., The authors are grateful to their universities. This paper has been supported by FEDER/Junta de Andalucía-Consejería de Transformación Económica, Industria, Conocimiento y Universidades by Project A-FQM-170-UGR20, and also by Ministerio de Ciencia e Innovación by Projects PID2020-119478GB-I00 and PID2019-108392GB-I00 (AEI/ 10.13039/501100011033).




Generalizing max pooling via (a, b)-grouping functions for convolutional neural networks

Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
  • 0000-0002-9960-0203
  • 0000-0002-7066-7156
  • 0000-0001-6986-9888
  • Herrera, Francisco
  • Takáč, Zdenko
  • 0000-0002-1279-6195
Due to their high adaptability to varied settings and effective optimization algorithm, Convolutional Neural
Networks (CNNs) have set the state-of-the-art on image processing jobs for the previous decade. CNNs work in
a sequential fashion, alternating between extracting significant features from an input image and aggregating
these features locally through ‘‘pooling" functions, in order to produce a more compact representation.
Functions like the arithmetic mean or, more typically, the maximum are commonly used to perform
this downsampling operation. Despite the fact that many studies have been devoted to the development of
alternative pooling algorithms, in practice, ‘‘max-pooling" still equals or exceeds most of these possibilities,
and has become the standard for CNN construction.
In this paper we focus on the properties that make the maximum such an efficient solution in the context
of CNN feature downsampling and propose its replacement by grouping functions, a family of functions that
share those desirable properties. In order to adapt these functions to the context of CNNs, we present (𝑎��, 𝑏��)-
grouping functions, an extension of grouping functions to work with real valued data. We present different
construction methods for (𝑎, 𝑏)-grouping functions, and demonstrate their empirical applicability for replacing
max-pooling by using them to replace the pooling function of many well-known CNN architectures, finding
promising results., The authors gratefully acknowledge the financial support of Tracasa Instrumental (iTRACASA) and of the Gobierno de Navarra -
Departamento de Universidad, Innovación y Transformación Digital,
as well as that of the Spanish Ministry of Science (project PID2019-108392GB-I00 (AEI/10.13039/501100011033)) and the project
PC095-096 FUSIPROD. T. Asmus and G.P. Dimuro are supported by the
projects CNPq (301618/2019-4) and FAPERGS (19/2551-0001279-9).
F. Herrera is supported by the Andalusian Excellence project P18-FR4961. Z. Takáč is supported by grant VEGA 1/0267/21. Open access
funding provided by Universidad Pública de Navarra.