Object detection using hybridization of static and dynamic feature spaces and its exploitation by ensemble classification

This paper presents a learning mechanism based on hybridization of static and dynamic learning. Realizing the detection performances offered by the state-of-the-art deep learning techniques and the competitive performances offered by the conventional static learning techniques, we propose the idea of exploitation of the concatenated (parallel) hybridization of the static and dynamic learning-based feature spaces. This is contrary to the cascaded (series) hybridization topology in which the initial feature space (provided by the conventional, static, and handcrafted feature extraction technique) is explored using deep, dynamic, and automated learning technique. Consequently, the characteristics already suppressed by the conventional representation cannot be explored by the dynamic learning technique. Instead, the proposed technique combines the conventional static and deep dynamic representation in concatenated (parallel) topology to generate an information-rich hybrid feature space. Thus, this hybrid feature space may aggregate the good characteristics of both conventional and deep representations, which are then explored using an appropriate classification technique. We also hypothesize that ensemble classification may better exploit this parallel hybrid perspective of the feature spaces. For this purpose, pyramid histogram of oriented gradients-based static learning has been incorporated in conjunction with convolution neural network-based deep learning to produce concatenated hybrid feature space. This hybrid space is then explored with various state-of-the-art ensemble classification techniques. We have considered the publicly available INRIA person and Caltech pedestrian standard image datasets to assess the performance of the proposed hybrid learning system. Furthermore, McNemar’s test has been used to statistically validate the outperformance of the proposed technique over various contemporary techniques. The validated experimental results show that the employment of the proposed hybrid representation results in effective detection performance (an AUC of 0.9996 for INRIA person and 0.9985 for Caltech pedestrian datasets) as compared to the individual static and dynamic representations.

     

Secure sensors data acquisition and communication protection in eHealthcare: Review on the state of the art

Recent advances in hardware technology have led to the development of low cost, power efficient and more feature rich devices that are amongst the most critical parts of communication networks. These devices or sensors can now sense data with more accuracy, process it by themselves and send it to the neighboring node or the sink node. However, robust and reliable security mechanisms are not yet properly implemented on these sensors due to their limited energy and computation power. Sensors also play a very important role in eHealthcare systems where ubiquitous patient monitoring is performed. As data is generated from the sensor nodes, reliable, secure and attack-resistant data acquisition and transmission is important for an efficient eHealthcare systems. This survey focuses on security issues of sensors data acquisition and transmission protocols, describing their main security features and comparing them in the context of a secure eHealthcare system. A taxonomy of open issues and future challenges is also discussed with respect to specific security metrics described in the paper.     

A finite element study of the effect of the angle between two intersecting cylindrical shells subjected to internal pressure

The results of a study of the critical region of two intersecting cylindrical shells due to internal pressure loading are presented as a function of the angle between the two axes. The investigation was performed using the thin shell element of a three-dimensional finite element program. Three models with the angle between the axes of the two cylindrical shells equal to 30, 60, and 90°, were analyzed. In all of the three models, the diameter ratio of the two shells was 0.5283; the diameter to thickness ratio of the larger or main shell was 44.76, while the same ratio for the smaller or attached shell was 15.487. The results of these analyses show that the stresses in the critical intersection region are least when the two axes are perpendicular to each other; for other angular configurations, the stresses increase as the acute angle between the two axes decrease. This effect of inclination for pressure loading is just opposite of the effect found by authors [1] for external moments. In that case, for in-plane as well as out-ofplane moments, the stresses are larger for normally intersecting shells as compared to other angular configurations.     

Self‐Assembly of Ferromagnetic Organic–Inorganic Perovskite‐Like Films

Perovskite‐based organic–inorganic hybrids hold great potential as active layers in electronics or optoelectronics or as components of biosensors. However, many of these applications require thin films grown with good control over structure and thickness—a major challenge that needs to be addressed. The work presented here is an effort towards this goal and concerns the layer‐by‐layer deposition at ambient conditions of ferromagnetic organic–inorganic hybrids consisting of alternating CuCl₄‐octahedra and organic layers. The Langmuir‐Blodgett technique used to assemble these structures provides intrinsic control over the molecular organization and film thickness down to the molecular level. Magnetic characterization reveals that the coercive field for these thin films is larger than that for solution‐grown layered bulk crystals. The strategy presented here suggests a promising cost effective route to facilitate the excellently controlled growth of sophisticated materials on a wide variety of substrates that have properties relevant for the high density storage media and spintronic devices.     

Diffusion in metallic glasses

Rutherfordα-particles backscattering technique was employed for measurements of diffusion rates in metallic glasses. Effects of relaxation, crystallization and plastic deformation on diffusion rates were also investigated. It has been observed that the diffusion rates of a metallic solute are of the same orders of magnitude in both metal-metal and metal-metalloid glasses. A higher diffusivity is likely if there is a large difference between melting points of the solute and matrix. Relaxation has no effect on diffusion, however, diffusivity increases on crystallization. An increase in diffusivity is also observed on plastic deformation of metallic glass.     

TD-LTE:抢占无线全新商机带动全球发展势态

LTE是目前大部分电信运营商首选的下一代移动通信技术,已有100多家运营商承诺将于2010年开始部署LTE.不过,在非成对频段部署LTE的商机却常被忽略.     

Discrete‐time Lyapunov stable‐staggered estimator MIMO adaptive interference cancelation with experimental verification

Adaptive interference cancelation is of vital importance in a broad array of scientific and engineering disciplines. In this paper we develop a closed‐loop discrete‐time interference cancelation algorithm. The novel features of this algorithm are its ability to deal with multiple channels being affected by interferences with different frequency spectrums. Also we provide a proof of Lyapunov stability of closed‐loop system and asymptotically perfect interference cancelation for a class of interference signals. Furthermore, we introduce a new approach for updating the estimator through the use of staggered estimate. The goal of staggered estimation is to minimize the total number of estimates/calculations done within a time period while ensuring that there is no estimator aliasing. Finally, the proposed algorithm is implemented on an TMS320C6713 DSP Kit and an experimental verification is obtained. Copyright © 2009 John Wiley & Sons, Ltd.     

A new hybrid approach for the solution of nonconvex economic dispatch problem with valve-point effects

Economic dispatch (ED) generally formulated as convex problem using optimization techniques by approximating generator input/output characteristic curves of monotonically increasing nature results in an inaccurate dispatch. The genetic algorithm has previously been used for the solution of problem for economic dispatch but takes longer time to converge to near optimal results. The hybrid approach is one of the methodologies used to fine tune the near optimal results produced by GA. This paper proposes new hybrid approach to solve the ED problem by using the valve-point effect. The approach we propose combines the genetic algorithm (GA) with active power optimization (APO) based on the Newton's second order approach (NSO). The genetic algorithm acts as a global optimizer giving near optimal generation schedule, which becomes the input for generation buses in APO algorithm. This algorithm acting as local search technique dispatching the generated active power of units for minimization of cost and gives optimum generation schedule. Three machines 6-bus, IEEE 5-machines 14-bus, and IEEE 6-mchines 30-bus systems have been tested for validation of our approach. Results of the proposed scheme compared with results obtained from GA alone give significant improvements in the generation cost showing the promise of the proposed approach.     

Role of Nd-Ni on structural,spectral and dielectric properties of strontium-barium based nano-sized X-type ferrites

The influence of neodymium and nickel substitution on structural and dielectric parameters was investigated in strontium-barium X-type hexagonal ferrites having composition SrBaCu_2?xNi_xNd_yFe_28?yO₄₆ (x = 0, 0.2, 0.4, 0.6, 0.8, 1 and y = 0, 0.02, 0.04, 0.06, 0.08, 0.1). Sol-gel method was employed for synthesizing these hexagonal ferrites. The XRD plots of all studied materials which were annealed at 1250 °C show single phase characteristics. Lattice parameter ‘c’ increased as a consequence of larger radius of rare earth ion (Nd³⁺) as compared to (Fe³⁺), while lattice parameter ‘a’ showed very small variation. The cell volume was obtained in the range 2508.32–2523.75 (ų). The inclusion of Nd-Ni also affected X-ray density, bulk density and porosity. The FTIR spectroscopy indicated the particular absorption peaks of hexagonal ferrites and it was performed in the range of 500–700 cm^?1. On account of Nd-Ni doping, the dielectric constant, dielectric loss and AC-conductivity showed decreasing trend. The occupancy of Nd³⁺ ions at octahedral site impedes the valence alternation of Fe³⁺; therefore there was decrease in dielectric permittivity. Ac conductivity has been decreased from 9.14 to 6.49 (Ω cm)^?1 at frequency of 2.7 GHz. The Cole-Cole plots of synthesized materials noticeably revealed grain boundary contribution. The appearance of single semi-circle in impedance Cole-Cole graphs confirms the exceptional role of grain boundaries in the conduction process. The considerably lower dielectric parameters of investigated nano X-type ferrites propose their feasibility for high-frequency applications (phase shifters, dielectric resonators, stealth technology etc).     

Interaction of cetyltrimethylammonium bromide with drug in aqueous/ electrolyte solution: A combined conductometric and molecular dynamics method study

Interaction between beta-lactum antibiotic drug ciprofloxacin hydrochloride(CFH)and cationic surfactant cetyltrimethylammonium bromide(CTAB)was performed conductometrically in aqueous as well as in the occurrence of different salts(NaCl,KCl as well as NH_4Cl)over the temperature range of 298.15–323.15 K at the regular interval of 5 K.CFH drug has been suggested for the treatment of bacterial infections such as urinary tract infections and acute sinusitis.A clear critical micelle concentration(CMC)was obtained for pure CTAB as well as(CFH+CTAB)mixed systems.The decrease in CMC values of CTAB caused by the addition of CFH reveals the existence of the interaction between the components and therefore it is the indication of micelle formation at lower concentration of CTAB and their CMC values further decrease in attendance of salts.A nonlinear behavior in the CMC versus T plot was observed in all the cases.The ΔG_m~0 values are found to be negative in present study systems demonstrated the stability of the solution.The values of ΔH_m~0 and ΔS_m~0 reveal the existence of hydrophobic and electrostatic interactions between CFH and CTAB.The thermodynamic properties of transfer for the micellization were also evaluated and discussed in detail.Molecular dynamic simulation disclosed that environment of water and salts have impact on the hydrophobic interaction between CFH and CTAB.In water and salts,CTAB adopts spherical micelle in which charged hydrophilic groups are interacted with waters whereas hydrophobic tails form the core of the micelle.This hydrophobic core region is highly conserved and protected.In addition,micelle formation is more favorable in aqueous Na Cl solution than other solutions.