This paper studies the problem of plenoptic sampling in image-based rendering (IBR). From a spectral analysis light field signals and using theorem, we mathematically derive analytical functions to determine minimum rate for rendering. The support signal is bounded by maximum depths only, no matter how complicated might be because depth variations scene. obtained compacting replicas sampled within smallest interval. Given depths, reconstruction filter with an optimal constant can designed...

Demand response (DR) is very important in the future smart grid, aiming to encourage consumers reduce their demand during peak load hours. However, if binary decision variables are needed specify start-up time of a particular appliance, resulting mixed integer combinatorial problem general difficult solve. In this paper, we study versatile convex programming (CP) DR optimization framework for automatic management various household appliances home. particular, an <i...

This paper presents a new superpixel-based hand gesture recognition system based on novel superpixel earth mover's distance metric, together with Kinect depth camera. The and skeleton information from are effectively utilized to produce markerless extraction. shapes, corresponding textures depths represented in the form of superpixels, which retain overall shapes color gestures be recognized. Based this representation, (SP-EMD), is proposed measure dissimilarity between gestures. measurement...

We survey the techniques for image-based rendering (IBR) and compressing representations. Unlike traditional three-dimensional (3-D) computer graphics, in which 3-D geometry of scene is known, IBR render novel views directly from input images. can be classified into three categories according to how much geometric information used: without geometry, with implicit (i.e., correspondence), explicit (either approximate or accurate geometry). discuss characteristics these their representative...

With the promises of smart grids, power can be more efficiently and reliably generated, transmitted, consumed over conventional electricity systems. Through two-way flow information between suppliers consumers, grids also adapt readily to increased penetration renewable energy sources encourage users' participation in savings cooperation through demand-response (DR) mechanism. An important issue is therefore how manage DR reduce peak load hence future investment thermal generations...

A class of subspace-based methods for direction-of-arrival (DOA) estimation and tracking in the case uniform linear arrays (ULAs) with mutual coupling is proposed. By treating angularly-independent as angularly-dependent complex array gains, middle subarray found to have same gains. Using this property, a new way parameterizing steering vector proposed corresponding method joint DOAs matrix (MCM) using whole data derived based on subspace principle. Simulation results show that algorithm has...

This paper studies the problem of robust adaptive filtering in impulsive noise environment using a recursive least M-estimate algorithm (RLM). The RLM minimizes M-estimator-based cost function instead conventional mean square error (MSE). Previous work has showed that offers improved robustness to impulses over squares (RLS) algorithm. In this paper, and convergence behaviors under contaminated Gaussian model is analyzed. A lattice structure-based fast algorithm, called Huber Prior Error...

Usually, direction-of-arrival (DOA) estimators are derived under the assumption of uniform white noise, whose covariance matrix is a scaled identity matrix. However, in practice, noise can be nonuniform with an arbitrary unknown diagonal In this situation, performance DOA may deteriorated considerably if nonuniformity ignored. To tackle problem, iterative approaches to subspace estimation developed and corresponding subspace-based addressed. our proposed methods, signal first determined by...

This paper proposes two gradient-based adaptive algorithms, called the least mean M estimate and transform domain M-estimate (TLMM) for robust filtering in impulse noise. A M-estimator is used as objective function to suppress adverse effects of noise on filter weights. They have a computational complexity order O(N) can be viewed, respectively, generalization square transform-domain algorithms. method fur estimating required thresholds also given. Simulation results show that TLMM...

This paper proposes a new digital beamformer for uniform concentric circular arrays (UCCAs) having nearly frequency-invariant (FI) characteristics. The basic principle is to transform the received signals phase mode and remove frequency dependency of individual through use beamforming or compensation network. As result, far-field pattern array, which governed by set variable weights, electronically steerable, it approximately invariant over wider range frequencies than conventional (UCAs)....

One of the most important applications in multiview imaging (MVI) is development advanced immersive viewing or visualization systems using, for instance, 3DTV. With introduction TVs, it expected that a new age 3DTV will arrive near future. Image-based rendering (IBR) refers to collection techniques and representations allow 3-D scenes objects be visualized realistic way without full model reconstruction. IBR uses images as primary substrate. The potential photorealistic has tremendous...

A new method for direction finding with partly calibrated uniform linear arrays (ULAs) is presented. It based on the conventional estimation of signal parameters via rotational invariance techniques (ESPRIT) by modeling imperfections ULAs as gain and phase uncertainties. For a fully array, it reduces to ESPRIT algorithm. Moreover, direction-of-arrivals (DOAs), unknown gains, phases uncalibrated sensors can be estimated in closed form without performing spectral search. Hence, computationally...

Object detection and tracking are two fundamental tasks in multicamera surveillance. This paper proposes a framework for achieving these nonoverlapping multiple camera network. A new object algorithm using mean shift (MS) segmentation is introduced, occluded objects further separated with the help of depth information derived from stereo vision. The detected then tracked by novel Bayesian Kalman filter simplified Gaussian mixture (BKF-SGM). It employs (GM) representation state noise...

The sequential partial-update least mean square (S-LMS)-based algorithms are efficient methods for reducing the arithmetic complexity in adaptive system identification and other industrial informatics applications. They also attractive acoustic applications where long impulse responses encountered. A limitation of these is their degraded performances an impulsive noise environment. This paper proposes new robust counterparts S-LMS family based on M-estimation. proposed M-estimate (S-LMM)...

Breast tumor segmentation of ultrasound images provides valuable information tumors for early detection and diagnosis. Accurate is challenging due to low image contrast between areas interest; speckle noises, large inter-subject variations in shape size. This paper proposes a novel Multi-scale Dynamic Fusion Network (MDF-Net) breast segmentation. It employs two-stage end-to-end architecture with trunk sub-network multiscale feature selection structurally optimized refinement mitigating...

This paper proposes a recursive least M-estimate (RLM) algorithm for robust adaptive filtering in impulse noise.It employs an cost function, which is able to suppress the effect of impulses on filter weights.Simulation results showed that RLM performs better than conventional RLS, NRLS, and OSFKF algorithms when desired input signals are corrupted by impulses.Its initial convergence, steady-state error, computational complexity, robustness sudden system change comparable RLS presence...

The discrete cosine transform (2-D DCT) is based on a one-dimensional fast (1-D FCT) algorithm. Instead of computing the 2-D using row-column method, 1-D algorithm extended by means vector-radix approach. Derivation both sequence splitting and Kronecker matrix product method are discussed. approach has advantage that all underlying operations shown clearly, while representations more compact readily generalized to higher dimensions. bit reversal placed before recursive additions so can be...

A new design method is proposed for the power or shaped beam pattern synthesis problem of narrowband conformal arrays, where only magnitude response specified. The iteratively linearizes non-convex function to obtain a convex subproblem in variables, which can be solved optimally using second-order cone programming (SOCP). In addition, wide variety constraints such as lower bound incorporated. An efficient technique determining reasonably good initial guess also further improve reliability...

Published methods that employ a filter bank for compensating the timing and bandwidth mismatches of an M-channel time-interleaved analog-to-digital converter (TIADC) were developed based on fact each sub-ADC channel is downsampled version analog input. The output filtered in such way that, when all outputs are summed, aliasing components minimized. If has N coefficients, optimization coefficients requires computing inverse MN times matrix if weighted least squares (WLS) technique used as...

This letter proposes a new adaptive beamforming algorithm for uniform linear arrays (ULAs) with unknown mutual coupling. It is based on the fact that coupling matrix (MCM) of ULA can be approximated as banded symmetric Toeplitz between two sensor elements inversely related to their separation, and hence it negligible when they are separated by few wavelengths. Taking advantage this specific structure MCM, approach calibrate signal steering vector proposed. By incorporating improved estimate...

This paper studies the design and implementation of finite-impulse response (FIR) infinite-impulse (IIR) variable digital filters (VDFs), whose frequency characteristics can be controlled continuously by some control or tuning parameters. A least squares (LS) approach is proposed for FIR VDFs expressing impulse filter as a linear combination basis functions. It shown that optimal LS solution obtained solving system equations. By choosing functions piecewise polynomials, with larger range...

This work studies the design and multiplier-less realization of a new software radio receiver (SRR) with reduced system delay. It employs low-delay finite-impulse response (FIR) digital allpass filters to effectively reduce delay multistage decimators in SRRs. The optimal least-square minimax designs these FIR allpass-based are formulated as semi-definite programming (SDP) problem, which allows zero magnitude constraint at /spl omega/=/spl pi/ be incorporated readily additional linear matrix...

According to Wang, there are four different types of DCT (discrete cosine transform) and DST sine the computation these sinusoidal transforms can be reduced type-IV DCT. As algorithms involve sizes at stages they not so regular in structure. Lee has developed a fast transform (FCT) algorithm for DCT-III similar decimation-in-time (DIT) Cooley–Tukey Fourier (FFT) with A disadvantage this is that it involves division trigonometric coefficients may numerically unstable. Recently, Hou an DCT-II...

This paper aims to investigate the relationship between torque and muscle morphological change, which is derived from ultrasound image sequence termed as sonomyography (SMG), during isometric ramp contraction of rectus femoris (RF) muscle, further compare SMG with electromyography (EMG) mechanomyography (MMG), represent electrical mechanical activities muscle. Nine subjects performed knee up 90% maximal voluntary (MVC) at speeds 45, 22.5 15% MVC/s, EMG, MMG ultrasonography were...

The filtered-x LMS (FxLMS) algorithm has been widely used in active noise control (ANC) systems, where the secondary path is usually estimated online by injecting auxiliary noises. In such an ANC system, controller and secondary-path estimator are coupled with each other, which make it difficult to analyze performance of entire system. Therefore, a comprehensive analysis broadband systems not available currently our best knowledge. this paper, convergence behavior FxLMS modeling studied....