Human Action Recognition in Compressed Domain using PBL-McRBFN Approach

Large variations in human actions lead to major challenges in computer vision research. Several algorithms are designed to solve the challenges. Algorithms that stand apart, help in solving the challenge in addition to performing faster and efficient manner. In this paper, we propose a human cognition inspired projection based learning for person-independent human action recognition in the H.264/AVC compressed domain and demonstrate a PBL-McRBEN based approach to help take the machine learning algorithms to the next level. Here, we use gradient image based feature extraction process where the motion vectors and quantization parameters are extracted and these are studied temporally to form several Group of Pictures (GoP). The GoP is then considered individually for two different bench mark data sets and the results are classified using person independent human action recognition. The functional relationship is studied using Projection Based Learning algorithm of the Meta-cognitive Radial Basis Function Network (PBL-McRBFN) which has a cognitive and meta-cognitive component. The cognitive component is a radial basis function network while the Meta-Cognitive Component(MCC) employs self regulation. The McC emulates human cognition like learning to achieve better performance. Performance of the proposed approach can handle sparse information in compressed video domain and provides more accuracy than other pixel domain counterparts. Performance of the feature extraction process achieved more than 90% accuracy using the PTIL-McRBFN which catalyzes the speed of the proposed high speed action recognition algorithm. We have conducted twenty random trials to find the performance in GoP. The results are also compared with other well known classifiers in machine learning literature.

Deformation field evolution in indentation of a porous brittle solid

An experimental study of plane strain wedge indentation of a model porous brittle solid has been made to understand the effect of indentation parameters on the evolution of the deformation field and the accompanying volume change. A series of high-speed, high-resolution images of the indentation region and simultaneous measurements of load response were captured through the progression of the indentation process. Particle image velocimetry analysis of the images facilitated in situ measurement of the evolution of the resulting plastic zone in terms of incremental material displacement (velocity), strain rate, strain and volume change (e.g., local pore collapse). These measurements revealed initiation and propagation of flow localizations and fractures, as well as enabled estimate of volume changes occurring in the deformation zone. The results were directly compared with theoretical estimates of indentation pressure and deformation zone geometry and were used to validate a modified cavity expansion solution that incorporates effects of volume changes in the plastic zone. (C) 2015 Elsevier Ltd. All rights reserved.

Compressed domain human action recognition in H.264/AVC video streams

This paper discusses a novel high-speed approach for human action recognition in H.264/AVC compressed domain. The proposed algorithm utilizes cues from quantization parameters and motion vectors extracted from the compressed video sequence for feature extraction and further classification using Support Vector Machines (SVM). The ultimate goal of the proposed work is to portray a much faster algorithm than pixel domain counterparts, with comparable accuracy, utilizing only the sparse information from compressed video. Partial decoding rules out the complexity of full decoding, and minimizes computational load and memory usage, which can result in reduced hardware utilization and faster recognition results. The proposed approach can handle illumination changes, scale, and appearance variations, and is robust to outdoor as well as indoor testing scenarios. We have evaluated the performance of the proposed method on two benchmark action datasets and achieved more than 85 % accuracy. The proposed algorithm classifies actions with speed (> 2,000 fps) approximately 100 times faster than existing state-of-the-art pixel-domain algorithms.

Phase Inversion of Agitated Liquid-Liquid Dispersions in the Presence of Micrometer-Sized Particles

We have investigated the impact of partially wetting particles of tens of micrometers on inversion instability of agitated liquid liquid dispersions. Particles of this size can be easily separated from the exit streams to avoid downstream processing-related issues. The results show that the presence of hydrophilic particles in small quantities (volume fraction range of 2 X 10(-4) to 1.25 x 10(-2)) significantly decreases the dispersed phase fraction at which water-in-oil (w/o) dispersions invert but leaves the inversion of oil-in-water (o/w) dispersions nearly unaffected. The addition of the same particles after they are hydrophobized decreases the dispersed phase fraction at which o/w dispersions invert but leaves the inversion of w/o dispersions unaffected. These findings suggest an increased rate of coalescence of drops when particles wet drops preferentially and a marginal decrease when they wet the continuous phase preferentially. High-speed conductivity measurements on w/o dispersion show transient conduction of a few hundred milliseconds duration through voltage pulses. Close to the inversion point, voltage pulses appear at high frequency for even 7 cm separation between the electrodes. The presence of hydrophilic particles produces a nearly identical signal at a significantly lower dispersed phase fraction itself, close to the new lowered inversion point in the presence of particles. We propose formation of elongated domains of the conducting dispersed phase through a rapid coalescence-deformation-breakup process to explain the new observations. The voltage signal appears as a forerunner of inversion instability.

Cadmium Sulfide Nanocrystal Sensitized Vertically Aligned Titanium Dioxide Rods for Large Area Image Sensors on 3-D Substrates

We discuss here a semiconductors assembly comprising of titanium dioxide (TiO2) rods sensitized by cadmium sulfide (CdS) nanocrystals for potential applications in large area electronics on three dimensional (3-D) substrates. Vertically aligned TiO2 rods are grown on a substrate using a 150 degrees C process flow and then sensitized with CdS by SILAR method at room temperature. This structure forms an effective photoconductor as the photo-generated electrons are rapidly removed from the CdS via the TiO2 thereby permitting a hole rich CdS. Current-voltage characteristics are measured and models illustrate space charge limited photo-current as the mechanism of charge transport at moderate voltage bias. The stable assembly and high speed are achieved. The frequency response with a loading of 10 pF and 9 M Omega shows a half power frequency of 100 Hz. (C) 2015 The Electrochemical Society. All rights reserved.

Quantitative In Situ Analysis of Deformation in Sliding Metals: Effect of Initial Strain State

Using in situ, high-speed imaging of a hard wedge sliding against pure aluminum, and image analysis by particle image velocimetry, the deformation field in sliding is mapped at high resolution. This model system is representative of asperity contacts on engineered surfaces and die-workpiece contacts in deformation and machining processes. It is shown that large, uniform plastic strains of 1-5 can be imposed at the Al surface, up to depths of 500 mu m, under suitable sliding conditions. The spatial strain and strain rate distributions are significantly influenced by the initial deformation state of the Al, e.g., extent of work hardening, and sliding incidence angle. Uniform straining occurs only under conditions of steady laminar flow in the metal. Large pre-strains and higher sliding angles promote breakdown in laminar flow due to surface fold formation or flow localization in the form of shear bands, thus imposing limits on uniform straining by sliding. Avoidance of unsteady sliding conditions, and selection of parameters like sliding angle, thus provides a way to control the deformation field. Key characteristics of the sliding deformation such as strain and strain rate, laminar flow, folding and prow formation are well predicted by finite element simulation. The deformation field provides a quantitative basis for interpreting wear particle formation. Implications for engineering functionally graded surfaces, sliding wear and ductile failure in metals are discussed.

Comparison of Convective and Radiative Heating Modes on the Thermophysical Changes of a Cerium Nitrate Droplet

In this paper, we try to establish the equivalence or similarity in the thermal and physiochemical changes in precursor droplets (cerium nitrate) in convective and radiative fields. The radiative field is created through careful heating of the droplet using a monochromatic light source (CO2 laser). The equivalence is also established for different modes of convection like droplet injected into a high-speed flow and droplet experiencing a convective flow due to acoustic streaming (levitated) only. The thermophysical changes are studied in an aqueous cerium nitrate droplet, and the dissociation of cerium nitrate to ceria is modeled using modified Kramers' reaction rate formulation. It is observed that vaporization, species accumulation, and chemical characteristics obtained in a convectively heated droplet are retained in a radiatively heated droplet by careful adjustment of the laser intensity. The timescales and ceria yield match reasonably well for both the cases. It is also noted that similar conclusions are drawn in both levitated droplet and a nonlevitated droplet.

All Inorganic Spin-Coated Nanoparticle-Based Capacitive Memory Devices

We demonstrate all inorganic, robust, cost-effective, spin-coated, two-terminal capacitive memory metal-oxide nanoparticle-oxide-semiconductor devices with cadmium telluride nanoparticles sandwiched between aluminum oxide phosphate layers to form the dielectric memory stack. Using a novel high-speed circuit to decouple reading and writing, experimentally measured memory windows, programming voltages, retention times, and endurance are comparable with or better than the two-terminal memory devices realized using other fabrication techniques.

Evaluation and Minimization of Low-Order Harmonic Torque in Low-Switching-Frequency Inverter-Fed Induction Motor Drives

A low-order harmonic pulsating torque is a major concern in high-power drives, high-speed drives, and motor drives operating in an overmodulation region. This paper attempts to minimize the low-order harmonic torques in induction motor drives, operated at a low pulse number (i.e., a low ratio of switching frequency to fundamental frequency), through a frequency domain (FD) approach as well as a synchronous reference frame (SRF) based approach. This paper first investigates FD-based approximate elimination of harmonic torque as suggested by classical works. This is then extended into a procedure for minimization of low-order pulsating torque components in the FD, which is independent of machine parameters and mechanical load. Furthermore, an SRF-based optimal pulse width modulation (PWM) method is proposed to minimize the low-order harmonic torques, considering the motor parameters and load torque. The two optimal methods are evaluated and compared with sine-triangle (ST) PWM and selective harmonic elimination (SHE) PWM through simulations and experimental studies on a 3.7-kW induction motor drive. The SRF-based optimal PWM results in marginally better performance than the FD-based one. However, the selection of optimal switching angle for any modulation index (M) takes much longer in case of SRF than in case of the FD-based approach. The FD-based optimal solutions can be used as good starting solutions and/or to reasonably restrict the search space for optimal solutions in the SRF-based approach. Both of the FD-based and SRF-based optimal PWM methods reduce the low-order pulsating torque significantly, compared to ST PWM and SHE PWM, as shown by the simulation and experimental results.