@inproceedings{ghanem_skau_krim_clouse_sakla_2016, title={Non-parametric bounds on the nearest neighbor classification accuracy based on the Henze-Penrose metric}, DOI={10.1109/icip.2016.7532581}, abstractNote={Analysis procedures for higher-dimensional data are generally computationally costly; thereby justifying the high research interest in the area. Entropy-based divergence measures have proven their effectiveness in many areas of computer vision and pattern recognition. However, the complexity of their implementation might be prohibitive in resource-limited applications, as they require estimates of probability densities which are very difficult to compute directly for high-dimensional data. In this paper, we investigate the usage of a non-parametric distribution-free metric, known as the Henze-Penrose test statistic, to estimate the divergence between different classes of vehicles. In this regard, we apply some common feature extraction techniques to further characterize the distributional separation relative to the original data. Moreover, we employ the Henze-Penrose metric to obtain bounds for the Nearest Neighbor (NN) classification accuracy. Simulation results demonstrate the effectiveness and the reliability of this metric in estimating the inter-class separability. In addition, the proposed bounds are exploited for selecting the least number of features that would retain sufficient discriminative information.}, booktitle={2016 ieee international conference on image processing (icip)}, author={Ghanem, S. and Skau, E. and Krim, H. and Clouse, H. S. and Sakla, W.}, year={2016}, pages={1364–1368} }
@inproceedings{skau_wohlberg_krim_dai_2016, title={Pansharpening via coupled triple factorization dictionary learning}, DOI={10.1109/icassp.2016.7471873}, abstractNote={Data fusion is the operation of integrating data from different modalities to construct a single consistent representation. This paper proposes variations of coupled dictionary learning through an additional factorization. One variation of this model is applicable to the pansharpening data fusion problem. Real world pansharpening data was applied to train and test our proposed formulation. The results demonstrate that the data fusion model can successfully be applied to the pan-sharpening problem.}, booktitle={International conference on acoustics speech and signal processing}, author={Skau, E. and Wohlberg, B. and Krim, H. and Dai, L. Y.}, year={2016}, pages={1234–1237} }
@article{stevens_skau_downen_roman_clarke_2011, title={Finite-size effects in nanocomposite thin films and fibers}, volume={84}, ISSN={["1550-2376"]}, DOI={10.1103/physreve.84.021126}, abstractNote={Monte Carlo simulations of finite-size effects for continuum percolation in three-dimensional, rectangular sample spaces filled with spherical particles were performed. For samples with any dimension less than 10-20 times the particle diameter, finite-size effects were observed. For thin films in the finite-size regime, percolation across the thin direction of the film gave critical volume fraction (p(c)) values that differed from those along the plane of the film. Simulations perpendicular to the film for very thin samples resulted in p(c) values lower than the classical limit of ∼29% (for spheres in a three-dimensional matrix) which increased with film thickness. For percolation along thin films, while holding film thickness constant, p(c) increased with increasing sample size, which is a modification of the finite-sized scaling effect for cubic samples. For samples with a large aspect ratio (fibers) and a finite-sized cross-sectional area, the critical volume fraction increased with sample length, as the sample became quasi-one-dimensional. The results are discussed in the context of adding volume along or perpendicular to the percolation direction. From an experimental perspective, these findings indicate that sample shape, as well as relative size, influences percolation in the finite-size regime.}, number={2}, journal={PHYSICAL REVIEW E}, author={Stevens, D. R. and Skau, E. W. and Downen, L. N. and Roman, M. P. and Clarke, L. I.}, year={2011}, month={Aug} }