Engagement Detection Based on Analyzing Micro Body Gestures Using 3D CNN

This paper proposes a novel, efficient and affordable approach to detect the students’ engagement levels in an e-learning environment by using webcams. Our method analyzes spatiotemporal features of e-learners’ micro body gestures, which will be mapped to emotions and appropriate engagement states. The proposed engagement detection model uses a three-dimensional convolutional neural network to analyze both temporal and spatial information across video frames. We follow a transfer learning approach by using the C3D model that was trained on the Sports-1M dataset. The adopted C3D model was used based on two different approaches; as a feature extractor with linear classifiers and a classifier after applying fine-tuning to the pre-trained model. Our model was tested and its performance was evaluated and compared to the existing models. It proved its effectiveness and superiority over the other existing methods with an accuracy of 94%. The results of this work will contribute to the development of smart and interactive e-learning systems with adaptive responses based on users’ engagement levels.     

Blood vessel segmentation and extraction using H-minima method based on image processing techniques

Multimedia Tools and Applications - In this paper, the H-minima transform is used for blood vessel segmentation. The aim of this study is to get the high accuracy of blood vessel segmentation in...     

Detection and spatial analysis of selective logging with geometrically corrected Landsat images

The Brazilian Amazonian rain forests are under imminent threat of serious degradation and ultimately deforestation. Human activities such as selective logging are an important cause. Selectively logged locations are difficult to detect from medium-resolution Landsat images, due to their relatively small sizes and subtle spatial patterns. Spectral linear unmixing provides an effective tool for the purpose. The orientation of geometrically corrected images, however, artificially introduces zero-reflectance background pixels. These change the variance–covariance structure of the image bands and hinder the identification of pure endmembers. In this study, we compare image cropping and image rotation as two alternative approaches. Selectively logged forests were detected in northern Rondônia state, north-western Mato Grosso state and south-eastern Amazonas state in Brazil by applying spectral unmixing. The study shows that image rotation is a better approach as it preserves the image extent and thus provides information on forest degradation over a wider region. Spatial statistical analysis of the detected locations shows strong clustering within the study area. We conclude that the endmembers used in this study represent basic components of a degraded forest environment. As spectral unmixing of remote-sensing images avoids collection of field data, it may broadly be applied towards other Amazonian regions as well.     

Modular multi-level converter-based DFIG wind power system with MMC-STATCOM

ABSTRACT

Transmission and distribution conversion operation, voltage, current magnitude variation and time-to-time power flow change through grid are the major issues in power system. Variable wind energy system is more popular for power generation through conventional converters and hence harmonic reduction is very less in these converters. In this paper, 9?MW modular multi-level converter-based DFIG wind energy power system parameters are compared with 9?MW DFIG systems using conventional converter. DFIG is operated in four mode condition with proper back-to-back power electronic converters and controllers. In power electronics, the modular multi-level converters are more advanced for reducing transients, peak response oscillations and improving the efficiency of the system. A five-level MMC has been applied in back-to-back converter with phase disposition (PD) PWM technique for DFIG wind system of the same rotor side controller and grid side controller. A conventional converter-based DFIG wind energy system is replaced with five-level MMC-based system with level shift PD-PWM technique. The impact of MMC on DFIG wind energy system is reduced voltage and current harmonics fast settling time compared to conventional converter-based DFIG system. The above MMC DFIG power system voltage is balanced at fault condition with MMC-STATCOM.     

Automatic detection and severity classification of diabetic retinopathy

Multimedia Tools and Applications - Diabetic retinopathy (DR) is a leading cause of preventable blindness caused by damaged blood vessels in the eye, if not treated early on. The aim of this...     

Channel coordination with quantity discounts and/or cooperative advertising

We assess the interactive effects of two commonly used channel coordination mechanisms (quantity discounts (QDs) and cooperative advertising (CA)). We use a game-theoretic model and solve four non-cooperative games. In the first game, neither QDs nor CA is implemented. Cooperative advertising alone is offered in the second game, while quantity discounts alone are offered in the third game. In the fourth game, both QDs and CA are implemented. We obtain analytical solutions and compare equilibrium results across games to assess the effectiveness of CA (QDs) when implemented alone or jointly with QDs (CA). The main findings suggest that the profitability of each of these mechanisms is affected by whether the other is implemented or not in the channel. For example, while CA benefits the manufacturer when implemented alone, it can increase or decrease the manufacturer’s profit when added to QDs. Looking at which coordination mechanism is most effective when used alone, we find that both the manufacturer and the supply chain prefer QDs to CA. Finally, the retailer may not benefit from either one or both of these coordination mechanisms, especially if marketing efforts are not highly effective.     

Performance Evaluation of MWCNTs Reinforced Cement Mortar Composites using Natural and Commercial Surfactants

An efficient and promising approach for effectively dispersing multi-walled carbon nanotubes (MWCNTs) in cementitious composites has been investigated.The naturally occurring organic extracts from species of indigenously known ‘Keekar’ trees scattered along tropical and sub-tropical regions;is found as an exceptional replacement to the non-natural commercial surfactants.In the initial phase of investigation,ideal surfactant’s content required for efficient dispersion of MWCNTs in solution was de...     

Chemical Fingerprinting of Heat Stress Responses in the Leaves of Common Wheat by Fourier Transform Infrared Spectroscopy

Wheat (Triticum aestivum L.) is known to be negatively affected by heat stress, and its production is threatened by global warming, particularly in arid regions. Thus, efforts to better understand the molecular responses of wheat to heat stress are required. In the present study, Fourier transform infrared (FTIR) spectroscopy, coupled with chemometrics, was applied to develop a protocol that monitors chemical changes in common wheat under heat stress. Wheat plants at the three-leaf stage were subjected to heat stress at a 42 °C daily maximum temperature for 3 days, and this led to delayed growth in comparison to that of the control. Measurement of FTIR spectra and their principal component analysis showed partially overlapping features between heat-stressed and control leaves. In contrast, supervised machine learning through linear discriminant analysis (LDA) of the spectra demonstrated clear discrimination of heat-stressed leaves from the controls. Analysis of LDA loading suggested that several wavenumbers in the fingerprinting region (400–1800 cm⁻¹) contributed significantly to their discrimination. Novel spectrum-based biomarkers were developed using these discriminative wavenumbers that enabled the successful diagnosis of heat-stressed leaves. Overall, these observations demonstrate the versatility of FTIR-based chemical fingerprints for use in heat-stress profiling in wheat.     

Wet-Chemical Synthesis and Electrochemical Properties of Ce-Doped FeVO4 for Use as New Anode Material in Li-ion Batteries

Ce-doped FeVO₄ nanocomposites were successfully synthesized using reverse micro-emulsion route. Thermal and microstructural characteristics were comprehensively investigated by simultaneous thermal analysis, X-ray diffraction (XRD), scanning and transmission electron microscopy, energy-dispersive X-ray spectroscopy, Fourier transform infrared spectroscopy (FT-IR) and laser particle size analyzer. Moreover, as the anode material of lithium-ion batteries, the electrochemical properties were studied by galvanostatic charge and discharge tests and electrochemical impedance spectroscopy. The thermal analysis illustrated that the triclinic crystal structure of FeVO₄ nanoparticles is formed at about 520 °C, which is confirmed by XRD and FT-IR results. Furthermore, the microstructural analyses revealed more regular particles and high specific surface area for wet-chemical derived FeVO₄:Ce, which decreases the diffusion pathway of the lithium ions during the insertion/extraction process. The electrochemical measurements indicated that the electrode cycling performance and rate retention ability of Ce-doped FeVO₄ are better than those of pure FeVO₄ due to the expansion of the crystal lattice, which provided more lattice space for lithium intercalation and de-intercalation. Consequently, the as-prepared Ce-doped FeVO₄ with relatively high specific and reversible capacity, thermal stability and satisfactory cycling performance is a promising candidate for use as a lithium batteries anode material.