Face anti-spoofing (a.k.a. presentation attack detection) has recently emerged as an active topic with great significance for both academia and industry due to the rapidly increasing demand in user authentication on mobile phones, PCs, tablets, so on. Recently, numerous face spoofing detection schemes have been proposed based assumption that training testing samples are same domain terms of feature space marginal probability distribution. However, unlimited variations dominant conditions...

Radio-frequency (RF) hepatic ablation, offers an alternative method for the treatment of malignancies. We employed finite-element (FEM) analysis to determine tissue temperature distribution during RF ablation. constructed three-dimensional (3-D) thermal-electrical FEM models consisting a four-tine probe, tissue, and large blood vessel (10-mm diameter) located at different locations. simulated our analyses under temperature-controlled (90/spl deg/C) 8-min also present preliminary result from...

In this paper, we propose a novel accurate detection framework of demosaicing regularity from different source images. The proposed first reversely classifies the demosaiced samples into several categories and then estimates underlying formulas for each category based on partial second-order derivative correlation models, which detect both intrachannel cross-channel correlation. An expectation-maximization reverse classification scheme is used to iteratively resolve ambiguous axes in order...

This paper proposes a novel Integrated Oversampling (INOS) method that can handle highly imbalanced time series classification. We introduce an enhanced structure preserving oversampling (ESPO) technique and synergistically combine it with interpolation-based oversampling. ESPO is used to generate large percentage of the synthetic minority samples based on multivariate Gaussian distribution, by estimating covariance minority-class regularizing unreliable eigen spectrum. To protect key...

Rhea (http://www.ebi.ac.uk/rhea) is a comprehensive resource of expert-curated biochemical reactions. provides non-redundant set chemical transformations for use in broad spectrum applications, including metabolic network reconstruction and pathway inference. includes enzyme-catalyzed reactions (covering the IUBMB Enzyme Nomenclature list), transport spontaneously occurring are described using species from Chemical Entities Biological Interest ontology (ChEBI) stoichiometrically balanced...

Three terbium complexes Tb(tba-PMP)3(TPPO) (A), Tb(tba-PMP)3(H2O) (B), and Tb(tba-PMP)3(Phen) (C) (where tba-PMP, TPPO, Phen stand for β-diketone 1-phenyl-3-methyl-4-(tert-butylacetyl)-5-pyrazolone, triphenyl phosphine oxide, 1,10-phenanthroline, respectively) with different neutral ligands were synthesized characterized, the mechanism of how affect photoluminescence (PL) electroluminescence (EL) properties was studied. Experiments revealed ligand TPPO strongly complex PL intensity, enhanced...

With advances in image display technology, recapturing good-quality images from the high-fidelity artificial scenery on a LCD screen becomes possible. Such posts security threat, which allows forgery to bypass current forensic systems. In this paper, we first recapture some photos different screens by properly setting up environment and tuning controllable settings. perceptional study, find that such finely recaptured can hardly be identified human eyes. To prevent attack, propose set of...

We describe efficient electroluminescence from a terbium complex, tris-(1-phenyl-3-methyl-4isobutyryl-5-pyrozolone)-bis(triphenyl phosphine oxide) (PTT). The green-emitting material possesses much higher photoluminescence efficiency compared to the commonly used green light emitter, 8-hydroxyquinoline aluminum (ALQ). rarely observed emission hole transport layer, N, N′-bis (3-methylphenyl)- N′-diphenyl-benzidine (TPD) of device ITO/TPD/PTT/Al proves that PTT also is good electron...

This paper presents an integrated framework to enable using standard non-sequential machine learning tools for accurate multi-modal activity recognition. We develop a novel that contains simple pre- and post-classification strategies improve the overall performance. achieve this through class-imbalance correction on data structure preserving oversampling (SPO), leveraging sequential nature of sensory smoothing predicted label sequence classifier fusion, respectively. Through evaluation...

We propose a novel framework of using parsimonious statistical model, known as mixture Gaussian trees, for modeling the possibly multimodal minority class to solve problem imbalanced time-series classification. By exploiting fact that close-by time points are highly correlated due smoothness time-series, our model significantly reduces number covariance parameters be estimated from O(d(2)) O(Ld), where L is components and d dimensionality. Thus, particularly effective high-dimensional with...

We propose, in this paper, a novel edge-adaptive data hiding method for authenticating binary host images. Through establishing dense grid (EAG) along the object contours, we use simple image to show that EAG more efficiently selects good carrying pixel locations (DCPL) associated with “ l-shaped” patterns than block-based methods. Our employs dynamic system structure redesigned fundamental content adaptive processes (CAP) switch iteratively trace new contour segments and search DCPLs. By...

In this paper, we study a cold-start heterogeneous-devicelocalization problem. This problem is challenging, becauseit results in an extreme inductive transfer learning setting,where there only source domain data but no target do-main data. also underexplored. As notarget for calibration, aim to learn robustfeature representation from the domain. There islittle previous work on such robust feature task; besides, existing propos-als are both heuristic and inexpressive. our contribution,we...

Heart disease is a serious threat to human health. Accurate prediction very important for prevention and treatment. The purpose of this study establish more suitable model heart disease. Based on LightGBM, we have deeply integrated bootstrap sampling weighting technology. We repeatedly use multiple parameters LightGBM perform the original training set. Through process, not only obtain sub-models various subsets but also mine rich data features. In process cross-validation, weight coefficient...

X. Gao, H. Cao, L. Zhang, B. Y. Cao and C. Huang, J. Mater. Chem., 1999, 9, 1077 DOI: 10.1039/A900276F

This paper presents a novel structure preserving over sampling (SPO) technique for classifying imbalanced time series data. SPO generates synthetic minority samples based on multivariate Gaussian distribution by estimating the covariance of class and regularizing unreliable eigen spectrum. By main intelligently creating protective variances in trivial feature dimensions, expand effectively into void area data space without being too closely tied with existing minority-class samples....

We propose in this paper a novel technique to correlate statistical image noise features with three EXchangeable Image File format (EXIF) header for manipulation detection. By formulating each EXIF feature as weighted sum of selected using sequential floating forward selection, the weights are then solved least squares solution modeling correlation between intact and corresponding header. manipulations like brightness contrast adjustment can affect these lead enlarged numerical difference...

In this paper, we propose a novel manipulation detection framework for image patches using fusion procedure, called FusionBoost, in conjunction with accurately detected derivative correlation features. By first dividing all demosaiced samples of color into number categories, estimate their underlying demosaicing formulas based on partial models and extract several types The features are organized small subsets according to both the category feature type. For each subset, train lightweight...

The authors used the finite element method to study effect of radio-frequency (RF) catheter ablation on tissue heating and lesion formation at different intracardiac sites exposed regional blood velocities. They examined application RF current in temperature- power-controlled mode above beneath mitral valve annulus where velocities are high low respectively. found that for temperature-controlled ablation, more power was delivered maintain preset tip temperature local velocity than velocity....

The apparent resistance of a finite-thickness layer measured with four-electrode plunge probe depends on the electrode insertion depth, spacing, and thickness, as well resistivity ratio an underlying layer. A physical model consisting air, saline solution layer, agar simulates real situation measurement. represents whose is to be by probe, perturbing micropositioner controls depth four electrodes into solution. With semi-infinite-thickness standard, measurement results show decreasing...

Here we present a novel, end-point method using the dead-end-elimination and A* algorithms to efficiently accurately calculate change in free energy, enthalpy, configurational entropy of binding for ligand-receptor association reactions. We apply new approach series human immunodeficiency virus (HIV-1) protease inhibitors examine effect ensemble reranking has on relative accuracy as well evaluate role absolute ligand losses upon affinity differences structurally related inhibitors. Our...

This study investigated the flow effect on lesion formation during radio-frequency cardiac catheter ablation in temperature-controlled mode. The blood heart chambers carries heat away from endocardium by convection. cooling requires more power generator and causes a larger lesion. We set up system to simulate inside chamber. performed vitro bovine myocardium with three different rates (0 L/min, 1 L/min 3 L/min) two target temperatures (60 degrees C 80 C). During ablation, we also recorded...