Pasar al contenido principal

METODOS INFORMADOS PARA LA SINTESIS DE SEÑALES

TEC2017-84743-P

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D (Excelencia / Generación de Conocimiento)
Año convocatoria 2017
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 16
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Parallelization of an algorithm for automatic classification of medical data

  • García Mollá, Víctor Manuel|||0000-0003-4768-7367
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vidal Maciá, Antonio Manuel
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Safont Armero, Gonzalo
In this paper, we present the optimization and parallelization of a state-of-the-art algorithm for automatic classification, in order to perform real-time classification of clinical data. The parallelization has been carried out so that the algorithm can be used in real time in standard computers, or in high performance computing servers. The fastest versions have been obtained carrying out most of the computations in Graphics Processing Units (GPUs). The algorithms obtained have been tested in a case of automatic classification of electroencephalographic signals from patients., This work was supported by Spanish Administration (Ministerio de Economía y Competitividad) and European Union (FEDER) under grants TEC2014-58438-R and TEC2017-84743-P; and Generalitat Valenciana under grants PROMETEO II/2014/032 and PROMETEO II/2014/003.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A New Surrogating Algorithm by the Complex Graph Fourier Transform (CGFT)

  • Belda, Jordi
  • Safont Armero, Gonzalo
  • Parcheta, Zuzanna
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
[EN] The essential step of surrogating algorithms is phase randomizing the Fourier transform while preserving the original spectrum amplitude before computing the inverse Fourier transform. In this paper, we propose a new method which considers the graph Fourier transform. In this manner, much more flexibility is gained to define properties of the original graph signal which are to be preserved in the surrogates. The complex case is considered to allow unconstrained phase randomization in the transformed domain, hence we define a Hermitian Laplacian matrix that models the graph topology, whose eigenvectors form the basis of a complex graph Fourier transform. We have shown that the Hermitian Laplacian matrix may have negative eigenvalues. We also show in the paper that preserving the graph spectrum amplitude implies several invariances that can be controlled by the selected Hermitian Laplacian matrix. The interest of surrogating graph signals has been illustrated in the context of scarcity of instances in classifier training., This research was funded by the Spanish Administration and the European Union under grant TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Estimation of the Number of Endmembers in Hyperspectral Images Using Agglomerative Clustering

  • Prades Nebot, José|||0000-0003-4489-3704
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Safont Armero, Gonzalo
[EN] Many tasks in hyperspectral imaging, such as spectral unmixing and sub-pixel matching, require knowing how many substances or materials are present in the scene captured by a yperspectral image. In this paper, we present an algorithm that estimates the number of materials in the scene using agglomerative clustering. The algorithm is based on the assumption that a valid clustering of the image has one cluster for each different material. After reducing the dimensionality of the hyperspectral image, the proposed method obtains an initial clustering using K-means. In this stage, cluster densities are estimated using Independent Component Analysis. Based on the K-means result, a model-based agglomerative clustering is performed, which provides a hierarchy of clusterings. Finally, a validation algorithm selects a clustering of the hierarchy; the number of clusters it contains is the estimated number of materials. Besides estimating the number of endmembers, the proposed method can approximately obtain the endmember (or spectrum) of each material by computing the centroid of its corresponding cluster. We have tested the proposed method using several hyperspectral images. The results show that the proposed method obtains approximately the number of materials that these images contain., This work was supported by Spanish Administration (Ministerio de Ciencia, Innovacion y Universidades) and European Union (FEDER) under grant TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Vector Score Alpha Integration for Classifier Late Fusion

  • Safont Armero, Gonzalo
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] Alpha integration is a family of integrators that encompasses many classic fusion operators (e.g., mean, product, minimum, maximum) as particular cases. This paper proposes vector score integration (VSI), a new alpha integration method for late fusion of multiple classifiers considering the joint effect of all the classes of the multi-class problem. Theoretical derivations to optimize the parameters of VSI for achieving the minimum probability of error are provided. VSI was applied to two classification tasks using electroencephalographic signals. The first task was the automatic stage classification of a neuropsychological test performed by epileptic subjects and the second one was the classification of sleep stages from apnea patients. Four single classifiers (linear and quadratic discriminant analysis, naive Bayes, and random forest) and three competitive fusion methods were estimated for comparison: mean, majority voting, and separated score integration (SSI). SSI is based on alpha integration, but unlike the proposed method, it considers the scores from each class in isolation, not accounting for possible dependencies among scores corresponding to different classes. VSI was able to optimally combine the results from all the single classifiers, in terms of accuracy and kappa coefficient, and outperformed the results of the other fusion methods in both applications., This work was supported by the Spanish Administration and European Union under grant TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A proxy learning curve for the Bayes classifier

  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Vidal, Enrique|||0000-0003-4579-5196
[EN] In this paper, a theoretical learning curve is derived for the multi-class Bayes classifier. This curve fits general multivariate parametric models of the class-conditional probability density. The derivation uses a proxy approach based on analyzing the convergence of a statistic which is proportional to the posterior probability of the true class. By doing so, the curve depends only on the training set size and on the dimension of the feature vector; it does not depend on the model parameters. Essentially, the learning curve provides an estimate of the reduction in the excess of the probability of error that can be obtained by increasing the training set size. This makes it attractive in order to deal with the practical problems of defining appropriate training set sizes., Acknowledgments Grant TEC2017-84743-P funded by MCIN/AEI/10.13039/50110 0 011033 and by the European Union.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A Comparative Study on Recent Automatic Data Fusion Methods

  • Pereira-González, Luis Manuel
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] Automatic data fusion is an important field of machine learning that has been increasingly studied. The objective is to improve the classification performance from several individual classifiers in terms of accuracy and stability of the results. This paper presents a comparative study on recent data fusion methods. The fusion step can be applied at early and/or late stages of the classification procedure. Early fusion consists of combining features from different sources or domains to form the observation vector before the training of the individual classifiers. On the contrary, late fusion consists of combining the results from the individual classifiers after the testing stage. Late fusion has two setups, combination of the posterior probabilities (scores), which is called soft fusion, and combination of the decisions, which is called hard fusion. A theoretical analysis of the conditions for applying the three kinds of fusion (early, late, and late hard) is introduced. Thus, we propose a comparative analysis with different schemes of fusion, including weaknesses and strengths of the state-of-the-art methods studied from the following perspectives: sensors, features, scores, and decisions., This research was funded by the Ministry of Science and Innovation of the Government of Spain MCIN/AEI/10.13039/501100011033 under Grants PRE2018-085092 and TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A comparative analysis of early and late fusion for the multimodal two-class problem

  • Pereira-González, Luis Manuel
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] In this article we carry out a comparison between early (feature) and late (score) multimodal fusion, for the two-class problem. The comparison is made first from a general perspective, and then from a specific mathematical analysis. Thus, we deduce the error probability expressions for the uncorrelated and correlated multivariate Gaussian distribution, assuming perfect model knowledge (Bayes error rates). We also deduce the corresponding expressions when the model is to be learned from a finite training set, demonstrating its convergence to the Bayes error rates as the training set size goes to infinite. These expressions also demonstrates that early fusion is the best option with model knowledge, and that both early and late fusion degrade due to a finite training set. This degradation is showed to be greater for early fusion due to the dimensionality increase of the feature space, so, eventually, late fusion could be a better option in a practical setting. The mathematical analysis also suggests the convenience of using a, so called, convergence factor, to quantify if a training set size is appropriate for the error probability to be close enough to the Bayes error rate. Different simulated experiments have been made to verify the validity of the mathematical analysis, as well as its possible extension to non-Gaussian models., This work was supported in part by MCIN/AEI/10.13039/501100011033 under Grant PRE2018-085092, and in part by Universitat Politecnica de Valencia.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Generative Adversarial Networks and Markov Random Fields for oversampling very small training sets

  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Safont, Gonzalo
[EN] In this work, we propose a new method for oversampling the training set of a classifier, in a scenario of extreme scarcity of training data. It is based on two concepts: Generative Adversarial Networks (GAN) and vector Markov Random Field (vMRF). Thus, the generative block of GAN uses the vMRF model to synthesize surrogates by the Graph Fourier Transform. Then, the discriminative block implements a linear discriminant on features measuring clique similarities between the synthesized and the original instances. Both blocks iterate until the linear discriminant cannot discriminate the synthetic from the original instances. We have assessed the new method, called Generative Adversarial Network Synthesis for Oversampling (GANSO), with both simulated and real data in experiments where the classifier is to be trained with just 3 or 5 instances. The applications consisted of classification of stages of neuropsychological tests using electroencephalographic (EEG) and functional magnetic resonance imaging (fMRI) data and classification of sleep stages using electrocardiographic (ECG) data. We have verified that GANSO can effectively improve the classifier performance, while the benchmark method SMOTE is not appropriate to deal with such a small size of the training set., This work was supported by Spanish Administration (Ministerio de Ciencia, Innovacion y Universidades) and European Union (FEDER) under grant TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Multiclass Alpha Integration of Scores from Multiple Classifiers

  • Safont Armero, Gonzalo
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] Alpha integration methods have been used for integrating stochastic models and fusion in the context of detection (binary classification). Our work proposes separated score integration (SSI), a new method based on alpha integration to perform soft fusion of scores in multiclass classification problems, one of the most common problems in automatic classification. Theoretical derivation is presented to optimize the parameters of this method to achieve the least mean squared error (LMSE) or the mínimum probability of error (MPE). The proposed alpha integrationmethod was tested on several sets of simulated and real data. The first set of experiments used synthetic data to replicate a problem of automatic detection and classification of three types of ultrasonic pulses buried in noise (four-class classification). The second set of experiments analyzed two databases (one publicly available and one private) of real polysomnographic records from subjects with sleep disorders. These records were automatically staged in wake, rapid eye movement (REM) sleep, and non-REM sleep (three-class classification). Finally, the third set of experiments was performed on a publicly available database of single-channel real electroencephalographic data that included epileptic patients and healthy controls in five conditions (five-class classification). In all cases, alpha integration performed better than the considered single classifiers and classical fusion techniques., This work was supported by the Spanish Administration and European Union under grant TEC2017-84743-P and Generalitat Valenciana under grant PROMETEO II/2014/032.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Computing the Partial Correlation of ICA Models for Non-Gaussian Graph Signal Processing

  • Belda, Jordi
  • Safont Armero, Gonzalo
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
[EN] Conventional partial correlation coefficients (PCC) were extended to the non-Gaussian case, in particular to independent component analysis (ICA) models of the observed multivariate samples. Thus, the usual methods that define the pairwise connections of a graph from the precision matrix were correspondingly extended. The basic concept involved replacing the implicit linear estimation of conventional PCC with a nonlinear estimation (conditional mean) assuming ICA. Thus, it is better eliminated the correlation between a given pair of nodes induced by the rest of nodes, and hence the specific connectivity weights can be better estimated. Some synthetic and real data examples illustrate the approach in a graph signal processing context., This research was funded by Spanish Administration and European Union under grants TEC2014-58438-R and TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Multichannel dynamic modeling of non-Gaussian mixtures

  • Safont Armero, Gonzalo
  • Gomez, Enriqueta
  • Villanueva, Vicente
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] This paper presents a novel method that combines coupled hidden Markov models (HMM) and non Gaussian mixture models based on independent component analyzer mixture models (ICAMM). The proposed method models the joint behavior of a number of synchronized sequential independent component analyzer mixture models (SICAMM), thus we have named it generalized SICAMM (G-SICAMM). The generalization allows for flexible estimation of complex data densities, subspace classification, blind source separation, and accurate modeling of both local and global dynamic interactions. In this work, the structured result obtained by G-SICAMM was used in two ways: classification and interpretation. Classification performance was tested on an extensive number of simulations and a set of real electroencephalograms (EEG) from epileptic patients performing neuropsychological tests. G-SICAMM outperformed the following competitive methods: Gaussian mixture models, HMM, Coupled HMM, ICAMM, SICAMM, and a long short-term memory (LSTM) recurrent neural network. As for interpretation, the structured result returned by G-SICAMM on EEGs was mapped back onto the scalp, providing a set of brain activations. These activations were consistent with the physiological areas activated during the tests, thus proving the ability of the method to deal with different kind of data densities and changing non-stationary and non-linear brain dynamics. (C) 2019 Elsevier Ltd. All rights reserved., This work was supported by Spanish Administration (Ministerio de Economia y Competitividad) and European Union (FEDER) under grants TEC2014-58438-R and TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

On Training Road Surface Classifiers by Data Augmentation

  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Rodríguez, Alberto
  • Vargas, Nancy
[EN] It is demonstrated that data augmentation is a promising approach to reduce the size of the captured dataset required for training automatic road surface classifiers. The context is on-board systems for autonomous or semi-autonomous driving assistance: automatic power-assisted steering. Evidence is obtained by extensive experiments involving multiple captures from a 10-channel multisensor deployment: three channels from the accelerometer (acceleration in the X, Y, and Z axes); three microphone channels; two speed channels; and the torque and position of the handwheel. These captures were made under different settings: three worm-gear interface configurations; hands on or off the wheel; vehicle speed (constant speed of 10, 15,20, 30 km/h, or accelerating from 0 to 30 km/h); and road surface (smooth flat asphalt, stripes, or cobblestones). It has been demonstrated in the experiments that data augmentation allows a reduction by an approximate factor of 1.5 in the size of the captured training dataset., This research was funded by MCIN/AEI/10.13039/501100011033 and by the European Union, grant number TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

New applications of late fusion methods for EEG signal processing

  • Safont, Gonzalo
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] Decision fusion consists in the combination of the outputs of multiple classifiers into a common decision that is more precise or stable. In most cases, however, only classical fusion techniques are considered. This work compares the performance of several state-of-the-art fusion methods on new applications of automatic stage classification of several neuropsychological tests. The tests were staged into three classes: stimulus display, retention interval, and subject response. The considered late fusion methods were: alpha integration; copulas; Dempster-Shafer combination; independent component analysis mixture models; and behavior knowledge space. Late fusion was able to improve the performance for the task, with alpha integration yielding the most stable result., This work was supported by Generalitat Valenciana under grant PROMETEO/2019/109 and Spanish Administration and European Union grant TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

A new method for fraud detection in credit cards based on transaction dynamics in subspaces

  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Safont, Gonzalo
[EN] This paper presents a new method for fraud detection in credit cards based on exploiting the dynamics of the card transactions. We hypothesize different behavior models in the use of the card between legitimate clients and fraudsters that are registered in the sequential pattern that follows the transactions. The method considers analyses in subspaces defined by two or three variables recorded in the transactions. From these subspaces, several dynamic features, such as transaction velocity and acceleration, are estimated as input vectors for a classification process. Linear and quadratic discriminant analysis and random forest are implemented as single classifiers. All the single classification results obtained for each of the subspaces are late fused to obtain an overall result using alpha integration algorithm. The proposed method was evaluated using a subset of real data with a very low fraud to legitimate transaction ratio. We demonstrated that the temporal dependence of card transactions exploited in different subspaces and fused to give an overall result improves the detection accuracy of fraud detection in credit cards., This work was supported by Generalitat Valenciana under grant PROMETEO/2019/109, and Spanish Administration and European Union grant TEC2017-84743-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

A Multisensor System for Road Surface Identification

  • Safont, Gonzalo
  • Rodríguez, Alberto
  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
[EN] This work introduces a multisensor road surface identification system that considers features from four different kind of sensors: microphones, accelerometers, speed signals, and handwheel signals. Features are extracted separately from each sensor, joined together, and then filtered using feature selection before classification. The proposed system was tested on a set of signals extracted from a specially-converted passenger car driving on a closed course. Three types of road surfaces were considered: smooth flat asphalt, cobblestones, and stripes. Three classifiers were considered: linear discriminant analysis, support vector machines, and random forests. All the considered classifiers reached over 90% accuracy, with a maximum accuracy of 96.52% for RDF. These results show the potential of the proposed system for road surface identification., This work was supported by Spanish Administration and European Union grant TEC2017-84743-P, and Generalitat Valenciana under grant PROMETEO/2019/109.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Pattern recognition techniques for provenance classification of archaeological ceramics using ultrasounds

  • Salazar Afanador, Addisson|||0000-0001-5849-5104
  • Vergara Domínguez, Luís|||0000-0001-6803-4774
  • Vidal, Enrique|||0000-0003-4579-5196
  • Safont Armero, Gonzalo
[EN] This paper presents a novel application of pattern recognition to the provenance classification of archae- ological ceramics. This is a challenging problem for archaeologists, which involves assigning a making location to a fragment of archaeological pottery that was found along with other fragments of pieces made in different distant locations from the find. The pieces look very similar to each other and, often, other contextual information about the use of the pieces cannot be used due to the small size of the fragments. Current standard methods to solve this problem are limited since they are time consuming, require costly equipment, and can lead to the destruction of a part of the pieces. The proposed method overcome those limitations using non-destructive ultrasonic testing and incorporates versatile data anal- ysis through advanced pattern recognition techniques. Those techniques include the following: feature ranking, sample augmentation, semi-supervision based on active learning; and optimal fusion. This latter is based in the concept of alpha integration, which allows optimal fitting of the fusion model parameters. Different provenance classification problems are showcased: provenance classification of terra sigillata ce- ramic pieces from Aretina, Northern Italy and Sud-Gaul origins; and provenance classification of Iberian ceramic pieces from archaeological sites of Paterna, and Les Jovaes in Valencia, Spain. We demonstrate that the proposed fusion-based method achieves the best results, in terms of balanced classification ac- curacy and F1 score, compared with competitive methods like linear discriminant analysis, random forest, and support vector machine. Experiments for simulating small sample sizes and uncertainty in labeling of the pieces are included. In addition, the paper provides a design of a practical specialized device that could be used in different applications of archaeological ceramic classification., This work was supported by Spanish Administration and European Union under grant TEC2017-84743-P.