An efficient approach for facial action unit intensity detection using distance metric learning based on cosine similarity

Emotions of human beings are largely represented by facial expressions. Facial expressions, simple as well as complex, are well decoded by facial action units. Any facial expression can be detected and analyzed if facial action units are decoded well. In the presented work, an attempt has been made to detect facial action unit intensity by mapping the features based on their cosine similarity. Distance metric learning based on cosine similarity maps the data by learning a metric that measures orientation rather than magnitude. The motivation behind using cosine similarity is that change in facial expressions can be better represented by changes in orientation as compared to the magnitude. The features are applied to support vector machine for classification of various intensities of action units. Experimental results on the popularly accepted database such as DISFA database and UNBC McMaster shoulder pain database confirm the efficacy of the proposed approach.     

Customising cost function for optimising image quality measures performances

This article proposes two reduced reference variance/covariance-based image quality metrics using a neural network approach. The main contribution is that the proposed metrics are computationally simple and do not require the entire reference images to be calculated while still giving higher performance ranges than 18 other full-reference image quality metrics available in the literature. The first metric called error-based cost function is more accurate than most of the others, while the second metric called correlation-based cost function outperforms the others in terms of correlation and monotonicity. A comparative study has been conducted over three image quality databases including the LIVE (second release), TID2008 and CSIQ.     

从风险分析到成本预算谈网络安全的平衡策略

本文从具体的风险分析结果出发,建立成本预算的数学模型,讨论在可接受的风险程度之内,尽量降低成本的投入,保证网络安全建设的高效性与平衡性的策略。     

Direct evaluation of a cost function in online identification techniques

Algorithms are presented for the online evaluation of the cost or error function of discrete online identification techniques. Both a direct and a recursive algorithm are given.     

Extraction of a cyclostationary source using a new cost function without pre-whitening

The proposed method aims to extract a cyclostationary source, whose cyclic frequency is a priori known, from a set of additive mixtures. The other sources may be either stationary or cyclostationary as long as their cyclic frequencies are different from that of the source to be extracted. The method does not require pre-whitening and consists in minimizing a criterion based on stationary and cyclostationary second order statistics of the observations; this method is labeled as Second Order Cyclostationary Statistics Optimization Criterion (SOC²). The relevance of this criterion is proven theoretically in the general case of N sources by P sensors, with P≥N. Other properties of the algorithm such as its accuracy and its robustness against additive noise or strong interferences are studied through a set of simulations.     

Systematic approach to MUSIC estimation with minimisation of its cost function

Based on the minimisation of a cost function, a systematic approach to MUSIC estimation is proposed for the arrival angles of the signals impinging on a sensor array. The proposed method, which employs Brent's method for the task together with bisection bracketing, can significantly reduce the computational burden compared to the conventional algorithm for computing the whole spatial spectrum     

Unsupervised Bayesian convex deconvolution based on a field with an explicit partition function.

This paper proposes a non-Gaussian Markov field with a special feature: an explicit partition function. To the best of our knowledge, this is an original contribution. Moreover, the explicit expression of the partition function enables the development of an unsupervised edge-preserving convex deconvolution method. The method is fully Bayesian, and produces an estimate in the sense of the posterior mean, numerically calculated by means of a Monte-Carlo Markov chain technique. The approach is particularly effective and the computational practicability of the method is shown on a simple simulated example.     

Convergence of back-propagation in neural networks using a log-likelihood cost function

A variation of the back-propagation algorithm is described, using a log-likelihood cost function. Appropriate choices of learning parameters are discussed. An example is given where the range of initial weights leading to proper convergence is increased, and the number of iterations required is significantly reduced.<>     

Simultaneous estimation of first and second derivatives of a cost function

A correlation method for estimating the second derivative, as well as the first derivative, of a cost function which is quadratic in an input parameter is described. The method, using a 3-level msequence perturbation signal, enables the cost-function minimum to be reached in a single step in a noisefree system. The effects of noise, and the extension of the method to multiparameter cost functions, are discussed.     

Modeling and optimization of UWB communication networks through a flexible cost function

The traditional design of communication networks has rarely been able to focus on the optimization of global network properties. Ultra-wideband (UWB) radio is emerging as an attractive physical layer for wireless communication networks offering new opportunities for the principled design and optimization of network properties. We develop a framework for the principled design of UWB wireless networks based on a flexible cost function that can be tailored and scaled to a wide range of networks and applications, ranging from sensor networks to voice and data wireless networks. The function comprises cost terms associated with transmission, connection setup, interference, and quality-of-service. Multihop routing strategies are associated with admissible paths of minimal cost that are computable in linear time. The cost function together with the overall level of requests determine the dynamics of the connections and the equilibrium topology of the network. We report simulation results in the case of simple ring and square lattice networks.     

On computing the repair cost of a maintained reliability system with known repair cost function and probability of repair

In this paper, a simplified analytic cost model for maintained reliability system under opportunistic repair scheme is discussed. Life cycle cost curves under various operating life cycle times and linear repair cost function is derived.     

An efficient algorithm for finding zeros of a real function of twovariables

The procedure presented searches the interval in which a function changes sign and then automatically locks to a curve f(λ,p)=0, following it inside a given rectangular region. The method ensures that at each step a new pair of points with different function signs is generated, and, in effect, it minimizes the number of function evaluations. Complementary algorithms opening up opportunities for the further automation of the search process are also presented in outline. An example of the application of the new procedure is included. The proposed algorithm is particularly suitable for solving the electromagnetic problems leading to transcendent equations     

Radiation due to a convex curvature discontinuity of adielectric-coated perfect conductor

Surface waves radiate energy at discontinuities in the curvature of the guiding structure. By reciprocity, surface waves will be excited by plane waves incident upon such a discontinuity. Here, the problem of the radiation of a surface wave on a flat dielectric-coated perfect conductor incident upon an abrupt change to a dielectric-coated cylindrical conductor with a large radius of curvature is considered. The problem is formulated as an integral equation over the aperture of the discontinuity. Since the change in curvature is modest, an approximate perturbation solution to the integral equation is derived and the radiated field due to the discontinuity is found. This radiated field reduces to published results for an impedance surface approximation when that approximation is valid. The problem of mode conversion and associated radiation near higher mode cutoffs is also studied. It is found that near mode cutoffs, the higher order mode dominates the radiation pattern and causes the overall radiation pattern due to the discontinuity in curvature to be narrow and end fire. Away from cutoff, when all of the propagating bound modes are more tightly bound to the surface, the radiation pattern is less narrow and less end fire. For very tightly bound modes the pattern is nearly uniform. For dielectrics characterized by small permitivities, the changes in radiation pattern should be measurable     

A pulse compression predistortion function for efficient sidelobe reduction in a high-power radar

The successful application of pulse compression techniques depends largely on reduction of the range sidelobes associated with the compressed pulse waveform. It is shown that the calculation of sidelobe levels in previously published data is not exact when low sidelobes are called for, since the effects of spectrum amplitude ripple in contributing to the range sidelobes have either been ignored or not stated. This paper presents more recent information on this subject, introducing a model of the Fresnel spectrum that permits a more accurate calculation of the compressed pulse sidelobe structure. Additional spectrum-associated paired-echo range sidelobes are predicted as a result of the model, and are observed in practice, having peak values up to 15 db larger than those derived on the basis of the idealized spectrum assumed in earlier studies. It is shown experimentally that elimination of the spectrum amplitude ripple results in removal of these paired-echo sidelobes, thus allowing more efficient use of the signal energy when low side-lobe levels are required. A predistorted linear FM pulse compression signal is presented as a realizable means for exerting control over the spectrum function. This new pulse compression technique does not require pulse envelope or rise-time shaping at the transmitter, and thus has application to high-power systems. Experimental waveforms illustrate the reduction of the spectrum-associated paired-echo sidelobes with the modified linear FM signal, and also show that the remaining range sidelobes do not appreciably increase as a function of Doppler shift.     

On a computationally efficient approach to boron-interstitial clustering

The physical concepts developed to describe the transient activation of boron during post-implantation annealing are based on the concurrent formation of complexes comprising boron atoms and self-interstitials. A complete implementation into TCAD software leads to a high number of equations to be solved which is often inadmissible for multi-dimensional simulations. In this work, a minimum number of such complexes is taken into considerations. We show that such a model is nevertheless able to reproduce profile shape and dopant activation for a large variety of implant and annealing conditions.     

Model selection through a statistical analysis of the globalminimum of a weighted nonlinear least squares cost function

This paper presents a model selection algorithm for the identification of parametric models that are linear in the measurements. It is based on the mean and variance expressions of the global minimum of a weighted nonlinear least squares cost function. The method requires the knowledge of the noise covariance matrix but does not assume that the true model belongs to the model set. Unlike the traditional order estimation methods available in literature, the presented technique allows to detect undermodeling. The theory is illustrated by simulations on signal modeling and system identification problems and by one real measurement example     

High-performance pulse-width modulation AC/DC controller using novel under voltage lockout circuit according to Energy Star VI standard

This paper proposes a high-performance pulse-width modulation (PWM) AC/DC controller, which can drive a high-voltage (HV) 650-V power metal-oxide-semiconductor field-effect Transistor (MOSFET) in typical applications of adapters in portable electronic devices. In order to reduce the standby power consumption and improve the response speed in the start-up state, an improved under voltage lockout (UVLO) circuit without a voltage reference source or comparator is adopted. The AC/DC controller is fabricated using a 40-V 0.8-μm one-poly two-metal (1P2M) CMOS process, and it only occupies 1410 × 730 μm². A 12 V/2 A flyback topology for quick-charge application is illustrated as the test circuit, which is currently one of the most advanced power adapters in use. Test values show that the turn-on and the turn-off threshold voltages are 19.318 and 8.01 V, respectively. A high hysteresis voltage of 11.308 V causes the value of the power-charging capacitor to decrease to as low as 1 μF to reduce production cost. In addition, the start-up current of 2.3 μA is extremely small, and is attributed to a reduction in the system's standby power consumption. The final test results of the overall system are proven to meet the Energy Star VI standard. The controller has already been mass produced for industrial applications.     

Stochastic behaviour and profit function of a system with precautionary measures under abnormal weather

A repairable system under normal and abnormal weather conditions is analysed. The system can be in one of the three modes—normal, partial and total failure. Failure rates of system and rates of change of weather conditions are constant while the repair rates are arbitrary functions of time. The repair time distributions depend upon the state from which the repair starts and are invariant with the change of weather. Using regenerative point technique various reliability characteristics such as mean time to system failure, steady state availability, the probability that the repairman is busy, expected number of visits by repairman and expected profit earned by the system are obtained.     

Fitting maintenance cost vs maintainability data to a standard curve

Practical maintainability design problems involve discrete points of maintenance cost vs maintainability data. It is essential that these data be fitted to any of the standard cost vs maintainability curves. The conventional least squares method is used here to achieve this task.     

Study on the similarity of photonic crystal ring resonator cavity modes and whispering-gallery-like modes in order to design more efficient optical power dividers

In this study, we discuss the wave pattern characteristics of confined cavity modes inside photonic crystal ring resonators (PCRRs). Despite different physical origins, these cavity modes are analogous to the whispering-gallery (WG-like) modes. Because of the absence of perfectly circular symmetry in our PC cavity, the WG-like modes are not degenerate, but they form a close doublet in which the field pattern of each of these doublets repeats itself by \(180^{\circ }/m\) where “m” is the azimuthal index number of WG-like modes. The cavity modes are named according to their symmetric field pattern, and WG-like ones are named by their azimuthal mode numbers “m.” Based on the study of these similarities, we propose a \(1\times 2\) and a \(1\times 4\) T-junction- like power dividers. Through study of the similar properties of these cavity modes with WG-like modes, we have enhanced the output power of both dividers to 49.6 and 24.5 %, respectively. The permittivity of the hexagonal photonic crystal rod is \(\varepsilon _{h}= 12.04\) surrounded by air as the background medium. For our PCRR, the dominant resonating mode is the cavity mode with hexapole field pattern or equivalently WG-like mode with azimuthal mode number of \(m\,=\,6\). The normalized frequencies for this mode and its doublet are \(a/\lambda = 0.348066\) and 0.348301, respectively. The photonic band structure, PC waveguide guided mode and electric field patterns of the confined cavity modes inside the PCRR are calculated using the PWE method, and the transmission spectra are calculated by means of 2D-FDTD simulation method.