Post-Nonlinear Mixtures and Beyond

Although sources in general nonlinear mixturm arc not separable iising only statistical independence, a special and realistic case of nonlinear mixtnres, the post nonlinear (PNL) mixture is separable choosing a suited separating system. Then, a natural approach is based on the estimation of tho separating Bystem parameters by minimizing an indcpendence criterion, like estimated mwce mutual information. This class of methods requires higher (than 2) order statistics, and cannot separate Gaarsian sources. However, use of [weak) prior, like source temporal correlation or nonstationarity, leads to other source separation Jgw rithms, which are able to separate Gaussian sourra, and can even, for a few of them, works with second-order statistics. Recently, modeling time correlated s011rces by Markov models, we propose vcry efficient algorithms hmed on minimization of the conditional mutual information. Currently, using the prior of temporally correlated sources, we investigate the fesihility of inverting PNL mixtures with non-bijectiw non-liacarities, like quadratic functions. In this paper, we review the main ICA and BSS results for riunlinear mixtures, present PNL models and algorithms, and finish with advanced resutts using temporally correlated snu~sm

Initialisation of Nonlinearities for PNL and Wiener systems Inversion

This paper proposes a very fast method for blindly initial- izing a nonlinear mapping which transforms a sum of random variables. The method provides a surprisingly good approximation even when the basic assumption is not fully satis¯ed. The method can been used success- fully for initializing nonlinearity in post-nonlinear mixtures or in Wiener system inversion, for improving algorithm speed and convergence.

Blind channel deconvolution of real world signals using source separation techniques

In this paper we present a method for blind deconvolution of linear channels based on source separation techniques, for real word signals. This technique applied to blind deconvolution problems is based in exploiting not the spatial independence between signals but the temporal independence between samples of the signal. Our objective is to minimize the mutual information between samples of the output in order to retrieve the original signal. In order to make use of use this idea the input signal must be a non-Gaussian i.i.d. signal. Because most real world signals do not have this i.i.d. nature, we will need to preprocess the original signal before the transmission into the channel. Likewise we should assure that the transmitted signal has non-Gaussian statistics in order to achieve the correct function of the algorithm. The strategy used for this preprocessing will be presented in this paper. If the receiver has the inverse of the preprocess, the original signal can be reconstructed without the convolutive distortion.

Improving a leaves automatic recognition process using PCA

In this work we present a simulation of a recognition process with perimeter characterization of a simple plant leaves as a unique discriminating parameter. Data coding allowing for independence of leaves size and orientation may penalize performance recognition for some varieties. Border description sequences are then used, and Principal Component Analysis (PCA) is applied in order to study which is the best number of components for the classification task, implemented by means of a Support Vector Machine (SVM) System. Obtained results are satisfactory, and compared with [4] our system improves the recognition success, diminishing the variance at the same time.

Automatic Recognition of Leaves by Shape Detection Pre-Processing with Ica

In this work we present a simulation of a recognition process with perimeter characterization of a simple plant leaves as a unique discriminating parameter. Data coding allowing for independence of leaves size and orientation may penalize performance recognition for some varieties. Border description sequences are then used to characterize the leaves. Independent Component Analysis (ICA) is then applied in order to study which is the best number of components to be considered for the classification task, implemented by means of an Artificial Neural Network (ANN). Obtained results with ICA as a pre-processing tool are satisfactory, and compared with some references our system improves the recognition success up to 80.8% depending on the number of considered independent components.

Coherency and sharpness measures by using ICA algorithms. An investigation for Alzheimer’s disease discrimination

In this paper, we present a comprehensive study of different Independent Component Analysis (ICA) algorithms for the calculation of coherency and sharpness of electroencephalogram (EEG) signals, in order to investigate the possibility of early detection of Alzheimer’s disease (AD). We found that ICA algorithms can help in the artifact rejection and noise reduction, improving the discriminative property of features in high frequency bands (specially in high alpha and beta ranges). In addition to different ICA algorithms, the optimum number of selected components is investigated, in order to help decision processes for future works.

Investigation of ICA algorithms for feature extraction of EEG signals in discrimination of Alzheimer disease

In this paper we present a quantitative comparisons of different independent component analysis (ICA) algorithms in order to investigate their potential use in preprocessing (such as noise reduction and feature extraction) the electroencephalogram (EEG) data for early detection of Alzhemier disease (AD) or discrimination between AD (or mild cognitive impairment, MCI) and age-match control subjects.

Blind Inversion of Wiener Systems

A system in which a linear dynamic part is followed by a non linear memoryless distortion a Wiener system is blindly inverted This kind of systems can be modelised as a postnonlinear mixture and using some results about these mixtures an e cient algorithm is proposed Results in a hard situation are presented and illustrate the e ciency of this algorithm

Differences of Functional Connectivity Brain Network in Emotional Judgment

Using combined emotional stimuli, combining photos of faces and recording of voices, we investigated the neural dynamics of emotional judgment using scalp EEG recordings. Stimuli could be either combioned in a congruent, or a non-congruent way.. As many evidences show the major role of alpha in emotional processing, the alpha band was subjected to be analyzed. Analysis was performed by computing the synchronization of the EEGs and the conditions congruent vs. non-congruent were compared using statistical tools. The obtained results demonstrate that scalp EEG ccould be used as a tool to investigate the neural dynamics of emotional valence and discriminate various emotions (angry, happy and neutral stimuli).

Parameterization of written signatures based on EFD

In this work we propose a method to quantify written signatures from digitalized images based on the use of Elliptical Fourier Descriptors (EFD). As usually signatures are not represented as a closed contour, and being that a necessary condition in order to apply EFD, we have developed a method that represents the signatures by means of a set of closed contours. One of the advantages of this method is that it can reconstruct the original shape from all the coefficients, or an approximated shape from a reduced set of them finding the appropriate number of EFD coefficients required for preserving the important information in each application. EFD provides accurate frequency information, thus the use of EFD opens many possibilities. The method can be extended to represent other kind of shapes.

Empirical mode decomposition-based face recognition system

In this work we explore the multivariate empirical mode decomposition combined with a Neural Network classifier as technique for face recognition tasks. Images are simultaneously decomposed by means of EMD and then the distance between the modes of the image and the modes of the representative image of each class is calculated using three different distance measures. Then, a neural network is trained using 10- fold cross validation in order to derive a classifier. Preliminary results (over 98 % of classification rate) are satisfactory and will justify a deep investigation on how to apply mEMD for face recognition.

What determines the wavelength of self-organized shoreline sand waves?

Shoreline undulations extending into the bathymetric contours with a length scale larger than that of the rhythmic surf zone bars are referred to as shoreline sand waves. Many observed undulations along sandy coasts display a wavelength in the order 1-7 km. Several models that are based on the hypothesis that sand waves emerge from a morphodynamic instability in case of very oblique wave incidence predict this range of wavelengths. Here we investigate the physical reasons for the wavelength selection and the main parametric trends of the wavelength in case of sand waves arising from such instability. It is shown that the existence of a minimum wavelength depends on an interplay between three factors affecting littoral drift: (A) the angle of wave fronts relative to local shoreline, which tends to cause maximum transport at the downdrift flank of the sand wave, (B) the refractive energy spreading which tends to cause maximum transport at the updrift flank and (C) wave focusing (de-focusing) by the capes (bays), which tends to cause maximum transport at the crest or slightly downdrift of it. Processes A and C cause decay of the sand waves while process B causes their growth. For low incidence angles, B is very weak so that a rectilinear shoreline is stable. For large angles and long sand waves, B is dominant and causes the growth of sand waves. For large angles and short sand waves C is dominant and the sand waves decay. Thus, wavelength selection depends on process C, which essentially depends on shoreline curvature. The growth rate of very long sand waves is weak because the alongshore gradients in sediment transport decrease with the wavelength. This is why there is an optimum or dominant wavelength. It is found that sand wave wavelength scales with λ0/β where λ0 is the water wave wavelength in deep water and β is the mean bed slope from shore to the wave base.

Alzheimer disease diagnosis based on automatic spontaneous speech analysis

Alzheimer’s disease (AD) is the most prevalent form of progressive degenerative dementia and it has a high socio-economic impact in Western countries, therefore is one of the most active research areas today. Its diagnosis is sometimes made by excluding other dementias, and definitive confirmation must be done trough a post-mortem study of the brain tissue of the patient. The purpose of this paper is to contribute to im-provement of early diagnosis of AD and its degree of severity, from an automatic analysis performed by non-invasive intelligent methods. The methods selected in this case are Automatic Spontaneous Speech Analysis (ASSA) and Emotional Temperature (ET), that have the great advantage of being non invasive, low cost and without any side effects.

EEG signal analysis via a cleaning procedure based on multivariate empirical mode decomposition

Artifacts are present in most of the electroencephalography (EEG) recordings, making it difficult to interpret or analyze the data. In this paper a cleaning procedure based on a multivariate extension of empirical mode decomposition is used to improve the quality of the data. This is achieved by applying the cleaning method to raw EEG data. Then, a synchrony measure is applied on the raw and the clean data in order to compare the improvement of the classification rate. Two classifiers are used, linear discriminant analysis and neural networks. For both cases, the classification rate is improved about 20%.

Estudi de la contaminació per nitrats a les aigües subterrànies de Catalunya: l'Alt Penedès, el Baix Camp, el Baix Empordà, el Baix Penedès, la Garrotxa, el Maresme, Osona, el Pla de l'Estany, la Segarra, el Segrià, el Tarragonès i l'Urgell

La contaminació per nitrats de les aigües subterrànies és un dels problemes mediambientals i de salut pública que més afecten els nostres aqüífers. És per això que un dels eixos del nostre estudi és fer la màxima difusió de l’estat de la qualitat de l’aigua dels aqüífers de diferents àrees de Catalunya. En el present estudi es tracta la contaminació per nitrats a les aigües de 945 punts de mostreig (sobretot d’aigua procedent de pous, d’algunes mines i d’algunes fonts) de Catalunya, situats en 12 comarques diferents, totes elles amb algun o una gran part dels seus municipis englobats en una zona vulnerable. D’aquests punts de mostreig s’ha fet un recull de dades de la concentració de nitrats des de l’any 2003 fins el 2012 per tal d’estudiar-ne l’evolució i la incidència de la pluviometria, així com conèixer en quines comarques les variables externes: superfície agrària útil (SAU), explotacions agràries i cabana porcina i bovina poden haver incidit directament en la contaminació. També s’estudia fins a quin punt, aquest deteriorament, ha afectat la mineralització de l’aigua o afecta a llocs d’aigües més mineralitzades. A partir dels resultats obtinguts s’ha arribat a les conclusions que la concentració mitjana de nitrats a les aigües subterrànies en la majoria de comarques estudiades és elevada, superior als 50 mg/l que estableix la normativa vigent; que ha tendit a l’estabilitat en els últims deu anys; que afecta llocs d’aigües molt mineralitzades i que la pluviometria incideix notablement en la concentració de nitrats. S’ha observat que les variables externes superfície agrària útil (SAU), explotacions agràries i cabana porcina i bovina incideixen directament en la contaminació, excepte al Maresme. Finalment es presenten diferents usos alternatius als de l’aigua de consum per aquestes aigües contaminades amb nitrats i que han posat en pràctica diferents municipis de les comarques estudiades.