In this paper, an image authentication technique by embedding digital "watermarks" into images is proposed. Watermarking a for labeling pictures hiding secret information the images. Sophisticated watermark potential method to discourage unauthorized copying or attest origin of our approach, we embed watermarks with visually recognizable patterns selectively modifying middle-frequency parts image. Several variations proposed are addressed. The experimental results show that successfully...

In this paper, we propose a novel low-rank appearance model for removing rain streaks. Different from previous work, our method needs neither pixel detection nor time-consuming dictionary learning stage. Instead, as streaks usually reveal similar and repeated patterns on imaging scene, generalize matrix to tensor structure in order capture the spatio-temporally correlated With model, thus remove image/video (and also other high-order image structure) unified way. Our experimental results...

The success of research on matrix completion is evident in a variety real-world applications. Tensor completion, which high-order extension has also generated great deal interest recent years. Given tensor with incomplete entries, existing methods use either factorization or schemes to recover the missing parts. However, as number entries increases, may overfit model because incorrectly predefined ranks, while fail interpret factors. In this paper, we introduce novel concept: complete and...

Due to the popularity of JPEG as an image compression standard, ability detect tampering in images has become increasingly important. Tampering compressed often involves recompression and tends erase traces found uncompressed images. In this paper, we present a new technique discover caused by recompression. We assume all source are format propose formulate periodic characteristics both spatial transform domains. Using theoretical analysis, design robust detection approach which is able...

Prestressed concrete structures, designed to enhance the compressive strength of through internal pretension, are increasingly susceptible serviceability issues caused by rising live loads, material degradation, and environmental impacts. Strengthening or retrofitting offers a practical cost-effective alternative full replacement. This study investigated flexural strengthening prestressed T-beams in negative moment region using near-surface mounted (NSM) carbon-fiber-reinforced polymer...

Fusion-born α particles moving parallel to the magnetic field can resonate with toroidal Alfvén eigenmodes (TAE) leading anomalous α-orbit diffusion across α-loss boundaries in a tokamak. This is analyzed using Hamiltonian guiding center code orbit conjunction kinetic magnetohydrodynamics (MHD) eigenmode solving nova-k. Resonant single orbits are studied below and above threshold for stochasticity Monte Carlo randomized ensembles of alphas subjected finite amplitude time-dependent TAE...

Rain removal from a video is challenging problem and has been recently investigated extensively. Nevertheless, the of rain single image rarely studied in literature, where no temporal information among successive images can be exploited, making it more challenging. In this paper, to best our knowledge, we are first propose single-frame-based framework via properly formulating as an decomposition based on morphological component analysis (MCA). Instead directly applying conventional...

Remote photoplethysmography (rPPG) is a contactless method for heart rate (HR) estimation from face videos. In this paper, we propose to estimate rPPG signals directly input video sequences in an end-to-end manner. We novel Siamese-rPPG network simultaneously learn the heterogeneous and homogeneous features two facial regions. Furthermore, analyze temporal periodicity of signals, construct with 3D CNNs jointly train two-branch model under negative Pearson loss function. Experimental results...

An image authentication technique by embedding each with a signature so as to discourage unauthorized copying is proposed. The proposed could actually survive several kinds of processing and the JPEG lossy compression.

The Hamiltonian four-field model is a simplified description of nonlinear tokamak dynamics that allows for finite ion Larmor radius physics, as well other effects related to compressibility and electron adiabaticity. Much simpler than some previous descriptions the same it still preserves essential features underlying exact dynamics. In particular, because dynamical system conserves appropriate Casimir invariants, avoiding implicit, unphysical dissipation. Here derived interpreted, its...

Remote photoplethysmography (rPPG) is a non-contact method for heart rate (HR) estimation from facial videos. In this paper, we propose novel two-stream convolutional neural network remote HR estimation. We introduce feature extraction stream by adopting low-rank constraint to guide the learn robust representation. also develop complementary stream, rPPG extract reliable signals regions. After fusing two streams, unified and estimate simultaneously. Experimental results on COHFACE dataset...

High-n, toroidal Alfven eigenmodes (TAEs) are studied on the basis of a kinetic model that includes full thermal ion finite Larmor radius effects, trapped electron collisions and fast particle instability drive. Lower TAEs (KTAEs) shown to be non-existent in nearly collisionless limit. Like TAEs, upper KTAEs exist because effects dissipationless Dissipation stability both can treated perturbatively. However, owing their extended mode structure ballooning space, usually remain stable or...

This paper aims to extract baseball game highlights based on audio-motion integrated cues. In order better describe different audio and motion characteristics in highlights, we propose a novel representation method likelihood models. The proposed models measure the "likeliness" of low-level features set predefined types categories, respectively. Our experiments show that using is more robust than audio/motion highlight. With models, then construct an feature by symmetrically fusing Finally,...

Phase-space topology of energetic particles in tokamak plasma with arbitrary shape cross section is studied based upon the guiding center theory. Important phase-space boundaries such as prompt loss boundary, trapped passing and other between classes nonstandard orbits (e.g., pinch stagnation orbits) are studied. This information applied to study anomalous diffusion due finite amplitude Alfvén wave fluctuations particles. The separatrix circulating contributes dominantly losses.

Reduced fluid models have become important tools for studying the nonlinear dynamics of plasma in a large aspect-ratio tokamak. A self-consistent reduced model, with finite ion-gyroradius effects is presented. The model distinctive being correct to O(( ρi/a)2) and satisfying an exact, relatively simple energy conservation law.

Age is one of the important biometric traits for reinforcing identity authentication. The challenge facial age estimation mainly comes from two difficulties: (1) wide diversity visual appearance existing even within same group and (2) limited number labeled face images in real cases. Motivated by previous research on human cognition, beings can confidently rank relative ages images, we postulate that plays a more role than attributes. In this paper, assume ranks be characterized set ranking...

This paper develops a systematic method of optimally locating and sizing the compensation capacitors on distribution feeders by taking into account mutual coupling effect among phase conductors. The equivalent circuit models three-phase, two-phase single-phase conductors are applied so that could be included in computer simulation. optimal capacitor placement is solved minimizing objective function which determined combination system peak power loss cost, energy cost cost. Both unbalance...

A study of radial gradient-driven neoclassical transport fast ions, produced by an isotropic source in axisymmetric toroidal system, is presented. The governing equation a flux surface averaged Fokker–Planck that retains both the effects pitch angle scattering and drag. It solved utilizing eigenfunction technique conjunction with rigorous numerical treatment. results, which are valid for finite inverse-aspect-ratio ε, found to fall between previous limiting calculations [Nucl. Fusion 15, 359...

This paper presents a graph-theoretic approach for interactive region-based image retrieval. When dealing with matching problems, we use graphs to represent images, transform the region correspondence estimation problem into an inexact graph problem, and propose optimization technique derive solution. We then define distance in terms of estimated correspondence. In relevance feedback steps, correspondence, maximum likelihood method re-estimate ideal query measurement. Experimental results...

Since JPEG image format has been a popularly used compression standard, tampering detection in images now plays an important role. The artifacts introduced by lossy can be seen as inherent signature for compressed images. In this paper, we propose new approach to analyse the blocking periodicity by, 1) developing linearly dependency model of pixel differences, 2) constructing probability map each pixel's belonging model, and 3) finally extracting peak window from Fourier spectrum map. We...

This paper presents an online feature selection algorithm for video object tracking. Using the and background pixels from previous frame as training samples, we model problem finding a good subset of features to better classify in current frame. aims improve existing methods by taking correlation between into consideration. We propose use AdaBoost iteratively select one which best compensates previously selected features. features, then construct compound likelihood image, shows ability...