Addisson Salazar

Optimum detection of ultrasonic echoes applied to the analysis of the first layer of a restored dome

Optimum detection is applied to ultrasonic signals corrupted with significant levels of grain noise. The aim is to enhance the echoes produced by the interface between the first and second layers of a dome to obtain interface traces in echo pulse B-scan mode. This is useful information for the restorer before restoration of the dome paintings. Three optimum detectors are considered: matched filter, signal gating, and prewhitened signal gating. Assumed models and practical limitations of the three optimum detectors are considered. The results obtained in the dome analysis show that prewhitened signal gating outperforms the other two optimum detectors.

A general procedure for learning mixtures of independent component analyzers

This paper presents a new procedure for learning mixtures of independent component analyzers. The procedure includes non-parametric estimation of the source densities, supervised-unsupervised learning of the model parameters, incorporation of any independent component analysis (ICA) algorithm into the learning of the ICA mixtures, and estimation of residual dependencies after training for correction of the posterior probability of every class to the testing observation vector. We demonstrate the performance of the procedure in the classification of ICA mixtures of two, three, and four classes of synthetic data, and in the classification of defective materials, consisting of 3D finite element models and lab specimens, in non-destructive testing using the impact-echo technique. The application of the proposed posterior probability correction demonstrates an improvement in the classification accuracy. Semi-supervised learning shows that unlabeled data can degrade the performance of the classifier when they do not fit the generative model. Comparative results of the proposed method and standard ICA algorithms for blind source separation in one and multiple ICA data mixtures show the suitability of the non-parametric ICA mixture-based method for data modeling.

Fast communication: On including sequential dependence in ICA mixture models

We present in this communication a procedure to extend ICA mixture models (ICAMM) to the case of having sequential dependence in the feature observation record. We call it sequential ICAMM (SICAMM). We present the algorithm, essentially a sequential Bayes processor, which can be used to sequentially classify the input feature vector among a given set of possible classes. Estimates of the class-transition probabilities are used in conjunction with the classical ICAMM parameters: mixture matrices, centroids and source probability densities. Some simulations are presented to verify the improvement of SICAMM with respect to ICAMM. Moreover a real data case is considered: the computation of hypnograms to help in the diagnosis of sleep disorders. Both simulated and real data analysis suggest the potential interest of including sequential dependence in the implementation of an ICAMM classifier.

Combination of multiple detectors for EEG based biometric identification/authentication

The different structures of the brain of human beings produce spontaneous electroencephalographic (EEG) records that can be used to identify subjects. This paper presents a method for biometric authorization and identification based on EEG signals. The hardware uses a simple 2-signal electrode and a reference electrode configuration. The electrodes are positioned in such a way to be as unobtrusive as possible for the tested subject. Multiple features are extracted from the EEG signals that are processed by different classifiers. The system uses all the possible combinations between classifiers and features, fusing the best results. The fused decision improves the classification performance for even a small number of observation vectors. Results were obtained from a population of 50 subjects and 20 intruders, both in authentication and identification tasks. The system obtains an Equal Error Rate (EER) of 2.4% with only a few seconds for testing. The obtained performance measures are an improvement over the results of current EEG-based systems.

ICA mixtures applied to ultrasonic nondestructive classification of archaeological ceramics

We consider a classifier based on Independent Component Analysis Mixture Modelling (ICAMM) to model the feature joint-probability density. This classifier is applied to a challenging novel application: classification of archaeological ceramics. ICAMM gathers relevant characteristics that have general interest for material classification. It can deal with arbitrary forms of the underlying probability densities in the feature vector space as nonparametric methods can do. Mutual dependences among the features are modelled in a parametric form so that ICAMM can achieve good performance even with a training set of relatively small size, which is characteristic of parametric methods. Moreover, in the training stage, ICAMM can incorporate probabilistic semisupervision (PSS): labelling by an expert of a portion of the whole available training set of samples. These properties of ICAMM are well-suited for the problem considered: classification of ceramic pieces coming from four different periods, namely, Bronze Age, Iberian, Roman, and Middle Ages. A feature set is obtained from the processing of the ultrasonic signal that is recorded in through-transmission mode using an ad hoc device. A physical explanation of the results is obtained with comparison with classical methods used in archaeology. The results obtained demonstrate the promising potential of ICAMM for material classification.

Semi-blind source extraction of atrial activity by combining statistical and spectral features

Atrial fibrillation is the most common human arrhythmia. During atrial fibrillation episodes, the surface electrocardiogram contains the linear superposition of the atrial and ventricular rhythms in addition to other non-cardiac artifacts. Since these signals can be considered statistically independent, a Blind Source Separation (BSS) approach fits the problem properly. The signal that contains useful clinical information is the atrial one. We present a solution that focuses on the extraction of the atrial activity, enforcing simultaneously the statistical and temporal properties of the atrial signal. In addition, we propose the use of kurtosis as a parameter to measure the quality of the extraction. The algorithm is applied successfully to synthetic and real data. It improves the extraction of the atrial signal in comparison to other BSS methods, recovers only the interesting atrial rhythm using the information contained in all the leads and reduces the computational cost. The results obtained are shown to be highly satisfactory, with an average of 53.9% of spectral concentration, -0.04 of kurtosis value, 2.98 of ventricular residua and 4.77% of significant QRS residua over a database of thirty patients.

Fast communication: Detection of signals of unknown duration by multiple energy detectors

An extension of the classical energy detector is proposed to deal with the case of unknown signal duration. Multiple energy detectors are applied to partitions of the original observation interval; presence of signal is decided if at least one of the detectors is in favor of it. We have derived the corresponding probabilities of false alarm and detection for a particular strategy of successive segmentations of the original interval, thus obtaining a layered structure of energy detectors. One key problem is that individual decisions obtained from the multiple energy detectors are statistically dependent, thus complicating the derivation of the overall probabilities of detection and false alarm. ROC curves have been computed, showing significant improvements in detectability when there is a large mismatch between the duration of the observation interval and the actual duration of the signal. This can be especially interesting in the framework of novelty detection where specific parameters of the signal, like duration, are totally unknown.

Semi-Supervised Bayesian Classification of Materials with Impact-Echo Signals

The detection and identification of internal defects in a material require the use of some technology that translates the hidden interior damages into observable signals with different signature-defect correspondences. We apply impact-echo techniques for this purpose. The materials are classified according to their defective status (homogeneous, one defect or multiple defects) and kind of defect (hole or crack, passing through or not). Every specimen is impacted by a hammer, and the spectrum of the propagated wave is recorded. This spectrum is the input data to a Bayesian classifier that is based on the modeling of the conditional probabilities with a mixture of Gaussians. The parameters of the Gaussian mixtures and the class probabilities are estimated using an extended expectation-maximization algorithm. The advantage of our proposal is that it is flexible, since it obtains good results for a wide range of models even under little supervision; e.g., it obtains a harmonic average of precision and recall value of 92.38% given only a 10% supervision ratio. We test the method with real specimens made of aluminum alloy. The results show that the algorithm works very well. This technique could be applied in many industrial problems, such as the optimization of the marble cutting process.

Automatic credit card fraud detection based on non-linear signal processing

Fraud detection is a critical problem affecting large financial companies that has increased due to the growth in credit card transactions. This paper presents a new method for automatic detection of frauds in credit card transactions based on non-linear signal processing. The proposed method consists of the following stages: feature extraction, training and classification, decision fusion, and result presentation. Discriminant-based classifiers and an advanced non-Gaussian mixture classification method are employed to distinguish between legitimate and fraudulent transactions. The posterior probabilities produced by classifiers are fused by means of order statistical digital filters. Results from data mining of a large database of real transactions are presented. The feasibility of the proposed method is demonstrated for several datasets using parameters derived from receiver characteristic operating analysis and key performance indicators of the business.

ICA Model Applied to Multichannel Non-destructive Evaluation by Impact-Echo

This article presents an ICA model for applying in Non Destructive Testing by Impact-Echo. The approach consists in considering flaws inside the material as sources for blind separation using ICA. A material is excited by a hammer impact and a convolutive mixture is sensed by a multichannel system. Obtained information is used for classifying in defective or non defective material. Results based on simulation by finite element method are presented, including different defect geometry and location.