Ponencias (Teoría de la Señal y Comunicaciones)

URI permanente para esta colecciónhttps://hdl.handle.net/11441/11424

Examinar

Mostrando 1 - 20 de 50

Generalization of an Active Set Newton Algorithm with Alpha-Beta divergences for audio separation
(European Signal Processing Conference, EUSIPCO, 2021-08) Sarmiento Vega, María Auxiliadora; Durán Díaz, Iván; Fondón García, Irene; Cruces Álvarez, Sergio Antonio; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Ministerio de Economia, Industria y Competitividad (MINECO). España; European Commission (EC). Fondo Europeo de Desarrollo Regional (FEDER); Consejería de Economía y Conocimiento, Junta de Andalucía; Universidad de Sevilla. TIC246: Tecnologías de aprendizaje automático y procesado digital de la información
This article considers the decomposition of a nonnegative signal into a non-negative linear combination of the contributions of pre-specified atomic units, which are also nonnegative. This model, referred as compositional model, is evident in the time-frequency characterisations of audio signals, where the sound can be viewed as a blending of spectral patterns of the component sounds that are present simultaneously. The algorithm proposed in this article obtains the activation vector of the atoms through an Active-Set Newton algorithm that employ the Alpha-Beta-divergence between the observed signal and the decomposition. This divergence family has been proved to be more efficient than other more common divergences, such as the generic Kullback-Leibler divergence in various audio signal processing applications. We have evaluated the proposed algorithm in a signal separation application of polyphonic music.
On the Minimum Perimeter Criterion for Bounded Component Analysis
(Institute of Electrical and Electronics Engineers (IEEE), 2023-05-05) Cruces Álvarez, Sergio Antonio; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Ministerio de Ciencia e Innovación (MICIN). España; Agencia Estatal de Investigación. España; European Commission (EC). Fondo Europeo de Desarrollo Regional (FEDER); Junta de Andalucía; Universidad de Sevilla. TIC246: Tecnologías de aprendizaje automático y procesado digital de la información
Bounded component analysis refers to a set of unsupervised techniques for the identification and recovery of underlying components of observations whose support has clear boundaries. This contribution is focused on the statistical justification and practical implementation of the minimum perimeter criterion, a bounded component analysis technique that can be used for the extraction of complex signals. In particular, with the help of the isoperimetric inequality, we reveal the link of the robust optimization of a statistical risk function with this criterion. We also address the difficulties for its optimization since, at the separation solutions, the criterion is non-differentiable. A reformulation of its optimization is shown to lead to a practical implementation that, for sources with clear boundaries, outperforms the normalized MSE performance of the state-of-the-art blind separation methods, being even close to the optimal performance of supervised methods.
Sparse Regression of Power Amplifier Behavioral Models with a Stagewise Doubly Orthogonal Matching Pursuit
(Institute of Electrical and Electronics Engineers (IEEE), 2023-02-20) Becerra González, Juan Antonio; Nogales, Miguel; Marqués Valderrama, Elías; Madero Ayora, María José; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Ministerio de Ciencia e Innovación (MICIN). España; Agencia Estatal de Investigación. España; European Commission (EC). Fondo Europeo de Desarrollo Regional (FEDER); Universidad de Sevilla. TIC158: Sistemas de Radiocomunicación
This work presents the sparse regression of Volterra-based behavioral models and digital predistorters (DPDs) through a Stagewise Doubly Orthogonal Matching Pursuit (StDOMP). First, the general sparse regression framework is presented. The StDOMP is formulated by applying the stagewise strategy to the DOMP algorithm, i.e., allowing the technique to select a block of components in each iteration without the need to perform one iteration per atom. The block selection approach allows to reduce the computational complexity up to a certain point where the performance is expected to be degraded. Experimental results highlight the behavior of the technique in the predistortion of a class AB power amplifier (PA) working in its nonlinear regime, concluding that the StDOMP is able to reduce runtime while still holding the performance level with respect to the DOMP.
Complex kernels for proper complex-valued signals: a review
(TIC155: Tratamiento de Señales y Comunicaciones (IEEE), 2015-12) Boloix Tortosa, Rafael; Payán Somet, Francisco Javier; Arias de Reyna Domínguez, Eva María; Murillo Fuentes, Juan José; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Ministerio de Educación y Ciencia (MEC). España; Universidad de Sevilla. TIC155: Tratamiento de Señales y Comunicaciones
In this paper we investigate the conditions that complex kernels must satisfy for proper complex-valued signals. We study the structure that complex kernels for proper complex-valued signals must have. Also, we demonstrate that complex kernels that have been previously proposed and used in adaptive filtering of complex-valued signals assume that those signals are proper, i.e, they are not correlated with their complex conjugate. We provide an example of how a complex-valued kernel suitable for a particular model is designed, with a procedure that could help in other designs. The experiments included show the good behavior of the proposed kernel in the task of nonlinear channel equalization.
Blind adaptive channel estimation for OFDM systems
(Institute of Electrical and Electronics Engineers (IEEE), 2009-06) Boloix Tortosa, Rafael; Simois Tirado, Francisco José; Murillo Fuentes, Juan José; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Universidad de Sevilla. TIC-155: Tratamiento de señales y comunicaciones
In this paper, we face the problem of blind channel estimation in zero padding OFDM systems. Based on the assumption that the transmitted symbols are independent and identically distributed, we propose a blind adaptive estimation algorithm for minimum phase channels, based on Independent Component Analysis, that exploits the particular structure of the ZP-OFDM system model. Some included simulations show that this novel algorithm can yield comparable performance as the SVD methods with a reduced computational complexity.
Gaussian processes regressors for complex proper signals in digital communications
(Institute of Electrical and Electronics Engineers (IEEE), 2014-06) Boloix Tortosa, Rafael; Payán Somet, Francisco Javier; Murillo Fuentes, Juan José; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Universidad de Sevilla. TIC-155: Tratamiento de señales y comunicaciones
In this paper we develop the complex-valued version of the Gaussian processes for regression (GPR) for proper complex signals. This tool has proved to be useful in the nonlinear detection in digital communications in real valued models. GPRs can be cast as nonlinear MMSE where hyperparameters can be tuned optimizing a marginal likelihood (ML). This feature allows for a flexible kernel that can easily adapt either to a linear or nonlinear solution. We introduce the complex-valued form of the GPR, and develop it for the proper complex case. We also deal with the optimization of the ML. Some experiments included illustrate the good performance of the proposal.
A Virtual Reality Prototype as a Tool Against Verbal Abuse in Classrooms: A Multidisciplinary Approach
(Science and Technology Publications, Lda, 2024) Fondón García, Irene; Elena Pérez, María del Mar; Jiménez Lagares, Irene; Gaytán Guía, Susana Pilar; Universidad de Sevilla. Departamento de Fisiología; Universidad de Sevilla. Departamento de Psicología Evolutiva y de la Educación; Universidad de Sevilla. Departamento de Ingeniería Electrónica; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
The exploration of affective response and its vegetative correlate to verbal abuse (VA) is a relevant research area to prevent bullying at schools. Taking advantage of the possibilities that virtual reality offers regarding to immersion in non-real environments inducing feelings in the users, this paper presents a virtual reality application focused on the study of VA in a school context. The versatility of the proposed project is directly related to its applicability. It has been designed under the premises of the psycho-neural effect of VA. The tool is intended to be used under professional and parental supervision, to perform experiments regarding bulling awareness. In this first stage, the authors propose a prototype that will be optimized and upgraded in future versions.
Hybrid and customized approach in telemedicine systems: An unavoidable destination
(IOS Press, 2005) Prado-Velasco, Manuel ; Roa Romero, Laura María; Reina Tosina, Luis Javier; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Universidad de Sevilla. Departamento de Ingeniería Gráfica; Instituto de Salud Carlos III; Instituto de Salud Carlos III
Several important problems in the majority of industrialized countries have challenged the centralized and overburdened current model of healthcare. Telehealthcare systems are presented as a new paradigm, offering high expectations to provide effective solutions to this picture. With this paper we present a new methodological approach for telehealthcare systems that pursues the generation of clinical and physiological knowledge of the patient in a real time and personalized manner. This approach is based on a computational component, identified as patient physiological image (PPI), which is responsible for generating an image of the state of the patient and therapy devices. Three key issues of the proposed methodological approach are evaluated. With the objective to validate the capability of the PPI to determine the internal state of a patient, a digital simulation experiment over the mathematical model of a PPI is done. Numerical results are compared to those obtained by a validated mathematical model. Secondly, a laboratory prototype of a novel human physical activity monitor that follows the designed methodological approach will be tested, in order to evaluate the trade-off between processing capacity, portability, and cost-efficiency and power consumption, which are necessary to assure its compliance with the methodology. As a third key issue, the capability of our methodology to integrate physiological information belonging to different scales is analyzed. This is done by means of a case study related to the integration of the regulation of water function of AQP2 channels (genomic, proteomic and cellular levels) into a kidney collecting duct epithelium mathematical model of a PPI. The analysis and preliminary evaluation of the proposed telehealthcare methodological approach, featured by an advanced personalization of health assistance, have been satisfactory.
Ambient Assisted Living: A methodological approach
(IEEE (Institute of Electrical and Electronics Engineers), 2010) Estudillo Valderrama, Miguel Ángel; Roa Romero, Laura María; Reina Tosina, Luis Javier; Román Martínez, Isabel; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper, the most important challenges and trends related to the application of Ambient Assisted Living (AAL) methods and techniques to the social/healthcare context are discussed. In order to find out technical solutions to these challenges, the main methodological issues concerning the design of open and distributed architectures are analyzed. The objective is to improve the efficiency/cost ratio in the provision of social and healthcare services to citizens with special needs, through the application of new paradigms in the context of AAL environments. Finally, some results and conclusions regarding the proposed open architecture are illustrated for the case of a distributed biomedical sensor network designed by the authors following this methodology.
Experimental Setup and Protocol for Creating an EEG-signal Database for Emotion Analysis Using Virtual Reality Scenarios
(SciTePress, 2023) Marqués Valderrama, Elías; Sarmiento Vega, María Auxiliadora; Durán Díaz, Iván; Becerra González, Juan Antonio; Fondón García, Irene; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Fondón García, Irene; European Commission (EC). Fondo Europeo de Desarrollo Regional (FEDER); Ministerio de Ciencia, Innovación y Universidades (MICINN). España; Agencia Estatal de Investigación. España; Universidad de Sevilla. TIC246: Tecnologías de Aprendizaje Automático y Procesado Digital de la Información
Automatic emotion recognition systems aim to identify human emotions from physiological signals, voice, facial expression or even physical activity. Among these types of signals, the usefulness of signals from electroencephalography (EEG) should be highlighted. However, there are few publicly accessible EEG databases in which the induction of emotions is performed through virtual reality (VR) scenarios. Recent studies have shown that VR has great potential to evoke emotions in an effective and natural way within a laboratory environment. This work describes an experimental setup developed for the acquisition of EEG signals in which the induction of emotions is performed through a VR environment. Participants are introduced to the VR environment via head-mounted displays (HMD) and 14-channel EEG signals are collected. The experiments carried out with 12 participants (5 male and 7 female) are also detailed, with promising results, which allow us to think about the future development of our own dataset.
Detección Automática de Microaneurismas en Retinografías
(Universidad Carlos III de Madrid, 2010-11) Núñez Benjumea, Francisco José; Fondón García, Irene; Acha Piñero, Begoña; Serrano Gotarredona, María del Carmen; Jiménez Carmona, Soledad; Alemany Márquez, Pedro; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
La detección de microaneurismas (MA) en retinografías es esencial a la hora de realizar un diagnóstico precoz de la retinopatía diabética (RD). Con esta finalidad, se presenta una herramienta automática que, tras preprocesar la retinografía con métodos basados en intensidad y tras realizar una selección de semillas mediante una umbralización adaptativa, obtiene una serie de candidatos a MA mediante un proceso de crecimiento de regiones, a partir de las cuales se seleccionan los verdaderos MA mediante una clasificación con una red neuronal Fuzzy ARTMAP. La evaluación del algoritmo se ha realizado con una base de datos consistente en 53 retinografías con 256 MA marcados por un oftalmólogo experto. 42 (204 MA) de estas imágenes han sido empleadas para entrenar al clasificador sirviendo las 11 (52 MA) restantes para la fase de prueba. La presente propuesta obtiene una sensibilidad (S) de 78.85% y una media de 9 falsos positivos por imagen (FPpI).
Novel multi-step region growing algorithm for the segmentation of burn images
(2005-11) Serrano Gotarredona, María del Carmen; Acha Piñero, Begoña; Fondón García, Irene; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper, a novel region growing algorithm for segmenting burn wounds in digital photographs of the skin is presented. This segmentation step has been developed as part of a CAD tool for burn diagnosis. During the region growing the inclusion condition relies on a tolerance parameter, which is adaptively determined. It initially has a low value and this is recursively increased until a stop condition is reached. This condition is based on the analysis of the statistical distribution of the pixels within the grown region. The great advantage of the method is that it does not rely on fine-tune parameters. The algorithm has been tested with 44 images of burn wounds obtaining very good results. Its performance has been measured quantitatively by comparing it with a ground truth obtained from the manual segmentation of five experts.
Detection of microcalcifications in mammograms using 2D prediction filtering and a new statistical measure of the right tail weight
(2005-11) Acha Piñero, Begoña; Serrano Gotarredona, María del Carmen; Rangayyan, Rangaraj M.; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper, a new method to detect microcalcifications in mammograms is presented. The method is based on a candidate selection procedure which consists of a two-dimensional linear prediction adaptive filtering followed by a statistical parameter calculation developed by the authors and called tail ratio (TR). The parameter TR characterizes the presence of microcalcifications in a ROI, and is extracted from the local probability distribution within a small region surrounding each candidate. Afterward a group of new and previously published features are used to feed a neural network that classifies the candidates into microcalcification or non-microcalcification. The algorithm has been tested with 38 digitized mammograms obtaining a sensitivity of 0.93 for a positive predictive value of 0.88.
Detección automática de microaneurismas en angiogramas de retina
(Sociedad Española de Ingeniería Biomédica, 2002) Cordero García, José Manuel; Serrano Gotarredona, María del Carmen; Acha Piñero, Begoña; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
En este artículo se presenta un nuevo método de detección automática de microaneurismas en angiogramas de retina, con el objetivo de servir de ayuda al diagnóstico de la retinopatía diabética en su etapa más precoz. Las características de esta enfermedad aconsejan un pronto diagnóstico, para el que resultan idóneas estas pequeñas lesiones retinianas, que se manifiestan especialmente claras en los angiogramas fluoresceínicos. El algoritmo propuesto para su detección y recuento se basa en la compensación ante variaciones locales y globales de las características de las imágenes, en la morfología matemática particularizada en el filtrado Top-Hat y en la validación de los microaneurismas a partir de tres novedosos criterios: intensidad, contraste y forma, que se combinan de forma que, tras haberse probado en diversos casos, se obtiene una tasa de acierto que habitualmente supera el 80%. Asimismo, se ha desarrollado un interfaz gráfico amigable que facilita su uso por parte del médico.
Global pattern classification in dermoscopic images based on modelling
(Sociedad Española de Ingeniería Biomédica, 2014) Sáez Manzano, Aurora; Acha Piñero, Begoña; Serrano Gotarredona, María del Carmen; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper a model-based method of classificationof global patterns in dermoscopic images is proposed. Global patterns identification is included in the pattern analysisframework, the melanoma diagnosis method most used amongdermatologists. The modelling is performed in two senses: first adermoscopic image is modelled by a finite symmetric conditionalMarkov model applied to colour space and the estimatedparameters of this model are treated as features. In turn, thedistribution of these features are supposed that follow a Gaussian mixturemodel along a lesion. The classification is carried out by an image retrieval approachwith different distance metrics. A 78.44% success rate in average is achieved when globular, homogeneous, and reticularare classified and a 72.91% success rate when the multicomponent pattern is added.
Colorimetric calibration of images of human skin captured under hospital conditions
(2005) Serrano Gotarredona, María del Carmen; Acha Piñero, Begoña; Sangwine, Steve J.; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper we study the problem of acquiring colorimetrically-calibrated images of human skin under hospital conditions with uncontrolled multiple illuminants. The application domain is diagnosis of skin burns based on the measured colour. To give some control over illumination we have experimented with xenon flash in an attempt to dominate the ambient illumination. We have used a Macbeth DC Color Checker as a test target to make measurements of observed colour using a digital camera under various illumination conditions. We present an iterative colorimetric calibration algorithm which determines both a spatial image correction profile and a transformation matrix to convert measured pixel values into calibrated CIE XYZ values. We have obtained a colorimetrically calibrated burn photograph database by applying this method.
Perceptual color clustering for color image segmentation based on CIEDE2000 color distance
(2009) Acha Piñero, Begoña; Serrano Gotarredona, María del Carmen; Fondón García, Irene; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper, a novel technique for color clustering with application to color image segmentation is presented. Clustering is performed by applying the k-means algorithm in the L*a*b* color space. Nevertheless, Euclidean distance is not the metric chosen to measure distances, but CIEDE2000 color difference formula is applied instead. K-means algorithm performs iteratively the two following steps: assigning each pixel to the nearest centroid and updating the centroids so that the empirical quantization error is minimized. In this approach, in the first step, pixels are assigned to the nearest centroid according to the CIEDE2000 color distance. The minimization of the empirical quantization error when using CIEDE2000 involves finding an absolute minimum in a non-linear equation and, therefore, an analytical solution cannot be obtained. As a consequence, a heuristic method to update the centroids is proposed. The proposed algorithm has been compared with the traditional k-means clustering algorithm in the L*a*b* color space with the Euclidean distance. The Borsotti parameter was computed for 28 color images. The new version proposed outperformed the traditional one in all cases.
Color analysis of the reconstructed complex nipple-areola after a mastectomy
(2009) Sáez Manzano, Aurora; Serrano Gotarredona, María del Carmen; Acha Piñero, Begoña; Gómez Cía, Tomás; Sicilia Castro, Domingo; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones
In this paper a colour analysis of the reconstructed NAC (nipple-areola complex) after a mastectomy is performed. First, a colour segmentation algorithm based on the Live Wire method is proposed to separate the NAC from the rest of skin. And then the colour differences between the healthy and the reconstructed NAC are measured using colour-difference formulas recommended in CIE: CIELAB, CIE94 and CIEDE2000. The application domain is analyzing how the NAC is modified after applying a new reconstructed technique of areola-nipple complex, grafted after its cryopreservation. The analysis has been performed for 20 images, and good segmentation results have been obtained and quantitative colour difference in accordance with perceptual colour difference has been obtained.
Combining blind source extraction with joint approximate diagonalization: Thin algorithms for ICA
(2003) Cruces Álvarez, Sergio Antonio; Cichocki, Andrzej; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Comisión Interministerial de Ciencia y Tecnología (CICYT). España
In this paper a multivariate contrast function is proposed for the blind signal extraction of a subset of the indepen dent components from a linear mixture. This contrast com bines the robustness of the joint approximate diagonaliza tion techniques with the flexibility of the methods for blind signal extraction. Its maximization leads to hierarchical and simultaneous ICA extraction algorithms which are respec tively based on the thin QR and thin SVD factorizations. The interesting similarities and differences with other exist ing contrasts and algorithms are commented.
Globally convergent Newton algorithms for blind decorrelation
(2003) Cruces Álvarez, Sergio Antonio; Cichocki, Andrzej; Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones; Comisión Interministerial de Ciencia y Tecnología (CICYT). España
This paper presents novel Newton algorithms for the blind adaptive decorrelation of real and complex processes. They are globally convergent and exhibit an interesting relation ship with the natural gradient algorithm for blind decorre lation and the Goodall learning rule. Indeed, we show that these two later algorithms can be obtained from their New ton decorrelation versions when an exact matrix inversion is replaced by an iterative approximation to it.

Examinar

Envíos recientes