Tensor LRR and sparse coding-based subspace clustering

Subspace clustering groups a set of samples from a union of several linear subspaces into clusters, so that the samples in the same cluster are drawn from the same linear subspace. In the majority of the existing work on subspace clustering, clusters are built based on feature information, while sample correlations in their original spatial structure are simply ignored. Besides, original high-dimensional feature vector contains noisy/redundant information, and the time complexity grows exponentially with the number of dimensions. To address these issues, we propose a tensor low-rank representation (TLRR) and sparse coding-based (TLRRSC) subspace clustering method by simultaneously considering feature information and spatial structures. TLRR seeks the lowest rank representation over original spatial structures along all spatial directions. Sparse coding learns a dictionary along feature spaces, so that each sample can be represented by a few atoms of the learned dictionary. The affinity matrix used for spectral clustering is built from the joint similarities in both spatial and feature spaces. TLRRSC can well capture the global structure and inherent feature information of data, and provide a robust subspace segmentation from corrupted data. Experimental results on both synthetic and real-world data sets show that TLRRSC outperforms several established state-of-the-art methods.

Sparse density estimator with tunable kernels

A new sparse kernel density estimator with tunable kernels is introduced within a forward constrained regression framework whereby the nonnegative and summing-to-unity constraints of the mixing weights can easily be satisfied. Based on the minimum integrated square error criterion, a recursive algorithm is developed to select significant kernels one at time, and the kernel width of the selected kernel is then tuned using the gradient descent algorithm. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing very sparse kernel density estimators with competitive accuracy to existing kernel density estimators.

Sparse density estimation on multinomial manifold combining local component analysis

A new sparse kernel density estimator is introduced based on the minimum integrated square error criterion combining local component analysis for the finite mixture model. We start with a Parzen window estimator which has the Gaussian kernels with a common covariance matrix, the local component analysis is initially applied to find the covariance matrix using expectation maximization algorithm. Since the constraint on the mixing coefficients of a finite mixture model is on the multinomial manifold, we then use the well-known Riemannian trust-region algorithm to find the set of sparse mixing coefficients. The first and second order Riemannian geometry of the multinomial manifold are utilized in the Riemannian trust-region algorithm. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing sparse kernel density estimators with competitive accuracy to existing kernel density estimators.

Sparse block-Jacobi matrices with arbitrarily accurate Hausdorff dimension

We show that the Hausdorff dimension of the spectral measure of a class of deterministic, i.e. nonrandom, block-Jacobi matrices may be determined with any degree of precision, improving a result of Zlatos [Andrej Zlatos,. Sparse potentials with fractional Hausdorff dimension, J. Funct. Anal. 207 (2004) 216-252]. (C) 2010 Elsevier Inc. All rights reserved.

Automatic camera control in virtual environments augmented using multiple sparse videos

Automated virtual camera control has been widely used in animation and interactive virtual environments. We have developed a multiple sparse camera based free view video system prototype that allows users to control the position and orientation of a virtual camera, enabling the observation of a real scene in three dimensions (3D) from any desired viewpoint. Automatic camera control can be activated to follow selected objects by the user. Our method combines a simple geometric model of the scene composed of planes (virtual environment), augmented with visual information from the cameras and pre-computed tracking information of moving targets to generate novel perspective corrected 3D views of the virtual camera and moving objects. To achieve real-time rendering performance, view-dependent textured mapped billboards are used to render the moving objects at their correct locations and foreground masks are used to remove the moving objects from the projected video streams. The current prototype runs on a PC with a common graphics card and can generate virtual 2D views from three cameras of resolution 768 x 576 with several moving objects at about 11 fps. (C)2011 Elsevier Ltd. All rights reserved.

Sparse partition universal graphs for graphs of bounded degree

In 1983, Chvatal, Trotter and the two senior authors proved that for any Delta there exists a constant B such that, for any n, any 2-colouring of the edges of the complete graph K(N) with N >= Bn vertices yields a monochromatic copy of any graph H that has n vertices and maximum degree Delta. We prove that the complete graph may be replaced by a sparser graph G that has N vertices and O(N(2-1/Delta)log(1/Delta)N) edges, with N = [B`n] for some constant B` that depends only on Delta. Consequently, the so-called size-Ramsey number of any H with n vertices and maximum degree Delta is O(n(2-1/Delta)log(1/Delta)n) Our approach is based on random graphs; in fact, we show that the classical Erdos-Renyi random graph with the numerical parameters above satisfies a stronger partition property with high probability, namely, that any 2-colouring of its edges contains a monochromatic universal graph for the class of graphs on n vertices and maximum degree Delta. The main tool in our proof is the regularity method, adapted to a suitable sparse setting. The novel ingredient developed here is an embedding strategy that allows one to embed bounded degree graphs of linear order in certain pseudorandom graphs. Crucial to our proof is the fact that regularity is typically inherited at a scale that is much finer than the scale at which it is assumed. (C) 2011 Elsevier Inc. All rights reserved.

Understanding the critical distance in sparse distributed memory

Modelos de tomada de decisão necessitam refletir os aspectos da psi- cologia humana. Com este objetivo, este trabalho é baseado na Sparse Distributed Memory (SDM), um modelo psicologicamente e neuro- cientificamente plausível da memória humana, publicado por Pentti Kanerva, em 1988. O modelo de Kanerva possui um ponto crítico: um item de memória aquém deste ponto é rapidamente encontrado, e items além do ponto crítico não o são. Kanerva calculou este ponto para um caso especial com um seleto conjunto de parâmetros (fixos). Neste trabalho estendemos o conhecimento deste ponto crítico, através de simulações computacionais, e analisamos o comportamento desta “Critical Distance” sob diferentes cenários: em diferentes dimensões; em diferentes números de items armazenados na memória; e em diferentes números de armazenamento do item. Também é derivada uma função que, quando minimizada, determina o valor da “Critical Distance” de acordo com o estado da memória. Um objetivo secundário do trabalho é apresentar a SDM de forma simples e intuitiva para que pesquisadores de outras áreas possam imaginar como ela pode ajudá-los a entender e a resolver seus problemas.

Sparse 1D discrete Dirac operators I: Quantum transport

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Sparse one-dimensional discrete Dirac operators II: Spectral properties

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

A W-matrix methodology for solving sparse network equations on multiprocessor computers

This paper describes a methodology for solving efficiently the sparse network equations on multiprocessor computers. The methodology is based on the matrix inverse factors (W-matrix) approach to the direct solution phase of A(x) = b systems. A partitioning scheme of W-matrix , based on the leaf-nodes of the factorization path tree, is proposed. The methodology allows the performance of all the updating operations on vector b in parallel, within each partition, using a row-oriented processing. The approach takes advantage of the processing power of the individual processors. Performance results are presented and discussed.

Sparse arrays and image compounding techniques for non-destructive testing using guided acoustic waves

Piezoelectric array transducers applications are becoming usual in the ultrasonic non-destructive testing area. However, the number of elements can increase the system complexity, due to the necessity of multichannel circuitry and to the large amount of data to be processed. Synthetic aperture techniques, where one or few transmission and reception channels are necessary, and the data are post-processed, can be used to reduce the system complexity. Another possibility is to use sparse arrays instead of a full-populated array. In sparse arrays, there is a smaller number of elements and the interelement spacing is larger than half wavelength. In this work, results of ultrasonic inspection of an aluminum plate with artificial defects using guided acoustic waves and sparse arrays are presented. Synthetic aperture techniques are used to obtain a set of images that are then processed with an image compounding technique, which was previously evaluated only with full-populated arrays, in order to increase the resolution and contrast of the images. The results with sparse arrays are equivalent to the ones obtained with full-populated arrays in terms of resolution. Although there is an 8 dB contrast reduction when using sparse arrays, defect detection is preserved and there is the advantage of a reduction in the number of transducer elements and data volume. © 2013 Brazilian Society for Automatics - SBA.

The use of instantaneous phase for improving sparse arrays images

Sparse arrays have pitch larger than half-wavelength (lambda/2) and there is a reduced number of elements in comparison with a full-populated array. Consequently, there is a reduction in cost, data acquisition and processing. However, conventional beamforming techniques result in large side and grating lobes, and consequently in image artifacts. In this work the instantaneous phase of the signals is used in a beamforming technique instead of the instantaneous amplitudes to improve images obtained from sparse arrays configurations. A threshold based on a statistical analysis and the number of signals used for imaging is applied to each pixel, in order to determine if that pixel is related to a defect or not. Three sets of data are used to evaluate the technique, considering medical and non-destructive testing: a simulated point spread function, a medical phantom and an aluminum plate with 2 lambda-, 7 lambda- and lambda-pitch, respectively. The conventional amplitude image is superposed by the image improved by the instantaneous phase, increasing the reflectors detectability and reducing artifacts for all cases, as well as dead zone for the tested plate.

A hybrid method for the solution of a sparse power system matrices on vector computers

This paper describes a methodology for solving a linear system of equations on vector computer. The methodology combines direct and inverse factors. The decomposition and implementation of the direct solution in a CRAY Y-MPZE/232, and the performance results are discussed.