Search

Abstract: 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. [Copyright &y& Elsevier]

In this paper, we propose and analyze by means of simulations the use of surrogate data algorithms for blind detection of nonlinearities in multiple-echo ultrasonic signals. We assume a blind scheme so that no information about the input (emitted ultrasonic pulse) can be used. The metrics and equations that model some nonlinear situations are carefully reviewed. Also, closed form equations of the third-order metrics from a simplified second-order Volterra kernel are derived. Computer simulations show that the surrogate data technique is a potentially powerful tool for blind detection of nonlinearities in multiple-echo ultrasonic signals if adequate metrics are chosen. They also reveal interesting trade-offs among parameters that model ultrasonic systems and detection percentages. [ABSTRACT FROM AUTHOR]

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 hyperspectral 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. [ABSTRACT FROM AUTHOR]