Optimal input cross-power spectra in shake table testing of asymmetric structures

The study considers earthquake shake table testing of bending-torsion coupled structures under multi-component stationary random earthquake excitations. An experimental procedure to arrive at the optimal excitation cross-power spectral density (psd) functions which maximize/minimize the steady state variance of a chosen response variable is proposed. These optimal functions are shown to be derivable in terms of a set of system frequency response functions which could be measured experimentally without necessitating an idealized mathematical model to be postulated for the structure under study. The relationship between these optimized cross-psd functions to the most favourable/least favourable angle of incidence of seismic waves on the structure is noted. The optimal functions are also shown to be system dependent, mathematically the sharpest, and correspond to neither fully correlated motions nor independent motions. The proposed experimental procedure is demonstrated through shake table studies on two laboratory scale building frame models.

A coupled discriminative dictionary and transformation learning approach with applications to cross domain matching

Cross domain and cross-modal matching has many applications in the field of computer vision and pattern recognition. A few examples are heterogeneous face recognition, cross view action recognition, etc. This is a very challenging task since the data in two domains can differ significantly. In this work, we propose a coupled dictionary and transformation learning approach that models the relationship between the data in both domains. The approach learns a pair of transformation matrices that map the data in the two domains in such a manner that they share common sparse representations with respect to their own dictionaries in the transformed space. The dictionaries for the two domains are learnt in a coupled manner with an additional discriminative term to ensure improved recognition performance. The dictionaries and the transformation matrices are jointly updated in an iterative manner. The applicability of the proposed approach is illustrated by evaluating its performance on different challenging tasks: face recognition across pose, illumination and resolution, heterogeneous face recognition and cross view action recognition. Extensive experiments on five datasets namely, CMU-PIE, Multi-PIE, ChokePoint, HFB and IXMAS datasets and comparisons with several state-of-the-art approaches show the effectiveness of the proposed approach. (C) 2015 Elsevier B.V. All rights reserved.

A coupled discriminative dictionary and transformation learning approach with applications to cross domain matching

Cross domain and cross-modal matching has many applications in the field of computer vision and pattern recognition. A few examples are heterogeneous face recognition, cross view action recognition, etc. This is a very challenging task since the data in two domains can differ significantly. In this work, we propose a coupled dictionary and transformation learning approach that models the relationship between the data in both domains. The approach learns a pair of transformation matrices that map the data in the two domains in such a manner that they share common sparse representations with respect to their own dictionaries in the transformed space. The dictionaries for the two domains are learnt in a coupled manner with an additional discriminative term to ensure improved recognition performance. The dictionaries and the transformation matrices are jointly updated in an iterative manner. The applicability of the proposed approach is illustrated by evaluating its performance on different challenging tasks: face recognition across pose, illumination and resolution, heterogeneous face recognition and cross view action recognition. Extensive experiments on five datasets namely, CMU-PIE, Multi-PIE, ChokePoint, HFB and IXMAS datasets and comparisons with several state-of-the-art approaches show the effectiveness of the proposed approach. (C) 2015 Elsevier B.V. All rights reserved.

Three-dimensional plotted hydroxyapatite scaffolds with predefined architecture: comparison of stabilization by alginate cross-linking versus sintering

Scaffolds for bone tissue engineering are essentially characterized by porous three-dimensional structures with interconnected pores to facilitate the exchange of nutrients and removal of waste products from cells, thereby promoting cell proliferation in such engineered scaffolds. Although hydroxyapatite is widely being considered for bone tissue engineering applications due to its occurrence in the natural extracellular matrix of this tissue, limited reports are available on additive manufacturing of hydroxyapatite-based materials. In this perspective, hydroxyapatite-based three-dimensional porous scaffolds with two different binders (maltodextrin and sodium alginate) were fabricated using the extrusion method of three-dimensional plotting and the results were compared in reference to the structural properties of scaffolds processed via chemical stabilization and sintering routes, respectively. With the optimal processing conditions regarding to pH and viscosity of binder-loaded hydroxyapatite pastes, scaffolds with parallelepiped porous architecture having up to 74% porosity were fabricated. Interestingly, sintering of the as-plotted hydroxyapatite-sodium alginate (cross-linked with CaCl2 solution) scaffolds led to the formation of chlorapatite (Ca9.54P5.98O23.8Cl1.60(OH)(2.74)). Both the sintered scaffolds displayed progressive deformation and delayed fracture under compressive loading, with hydroxyapatite-alginate scaffolds exhibiting a higher compressive strength (9.5 +/- 0.5MPa) than hydroxyapatite-maltodextrin scaffolds (7.0 +/- 0.6MPa). The difference in properties is explained in terms of the phase assemblage and microstructure.

Three-Dimensional Propagation of the Transmitted Shock Wave in a Square Cross-Sectional Chamber

An investigation into the three-dimensional propagation of the transmitted shock wave in a square cross-section chamber was described in this paper, and the work was carried out numerically by solving the Euler equations with a dispersion-controlled scheme. Computational images were constructed from the density distribution of the transmitted shock wave discharging from the open end of the square shock tube and compared directly with holographic interferograms available for CFD validation. Two cases of the transmitted shock wave propagating at different Mach numbers in the same geometry were simulated. A special shock reflection system near the corner of the square cross-section chamber was observed, consisting of four shock waves: the transmitted shock wave, two reflection shock waves and a Mach stem. A contact surface may appear in the four-shock system when the transmitted shock wave becomes stronger. Both the secondary shock wave and the primary vortex loop are three-dimensional in the present case due to the non-uniform flow expansion behind the transmitted shock.

The cross-entropy method for blind multiuser detection

We consider the problem of blind multiuser detection. We adopt a Bayesian approach where unknown parameters are considered random and integrated out. Computing the maximum a posteriori estimate of the input data sequence requires solving a combinatorial optimization problem. We propose here to apply the Cross-Entropy method recently introduced by Rubinstein. The performance of cross-entropy is compared to Markov chain Monte Carlo. For similar Bit Error Rate performance, we demonstrate that Cross-Entropy outperforms a generic Markov chain Monte Carlo method in terms of operation time.

Cross-relaxation dynamics on anomalous saturation processes in low pressure supersonic cw HF chemical laser amplifier

It is assumed that both translational and rotational nonequilibrium cross-relaxations play a role simultaneoulsy in low pressure supersonic cw HF chemical laser amplifier. For two-type models of gas flow medium with laminar and turbulent flow diffusion mixing, the expressions of saturated gain spectrum are derived respectively, and the numerical calculations are performed as well. The numerical results show that turbulent flow diffusion mixing model is in the best agreement with the experimental result.

Cross-sectional structure of tetrahedral amorphous carbon thin films

A cross-sectional transmission electron microscope study of the low density layers at the surface and at the substrate-film interface of tetrahedral amorphous carbon (ta-C) films grown on (001) silicon substrates is presented. Spatially resolved electron energy loss spectroscopy is used to determine the bonding and composition of a tetrahedral amorphous carbon film with nanometre spatial resolution. For a ta-C film grown with a substrate bias of -300 V, an interfacial region approximately 5 nm wide is present in which the carbon is sp2 bonded and is mixed with silicon and oxygen from the substrate. An sp2 bonded layer observed at the surface of the film is 1.3 ± 0.3 nm thick and contains no detectable impurities. It is argued that the sp2 bonded surface layer is intrinsic to the growth process, but that the sp2 bonding in the interfacial layer at the substrate may be related to the presence of oxygen from the substrate.

Evaluation of the Interfacial Adhesion between Brittle Coating and Ductile Substrate by Cross-Secitional Indention

The cross-sectional indentation method is extended to evaluate the interfacial adhesion between brittle coating and ductile substrate. The experimental results on electroplated chromium coating/steel substrate show that the interfacial separation occurs due to the edge chipping of brittle coating. The corresponding models are established to elucidate interfacial separation processes. This work further highlights the advantages and potential of this novel indentation method.

Language model cross adaptation for LVCSR system combination

State-of-the-art large vocabulary continuous speech recognition (LVCSR) systems often combine outputs from multiple subsystems developed at different sites. Cross system adaptation can be used as an alternative to direct hypothesis level combination schemes such as ROVER. In normal cross adaptation it is assumed that useful diversity among systems exists only at acoustic level. However, complimentary features among complex LVCSR systems also manifest themselves in other layers of modelling hierarchy, e.g., subword and word level. It is thus interesting to also cross adapt language models (LM) to capture them. In this paper cross adaptation of multi-level LMs modelling both syllable and word sequences was investigated to improve LVCSR system combination. Significant error rate gains up to 6.7% rel. were obtained over ROVER and acoustic model only cross adaptation when combining 13 Chinese LVCSR subsystems used in the 2010 DARPA GALE evaluation. © 2010 ISCA.