Chemical composition and nutritional quality of tehineh (sesame butter)

The chemical composition and nutritional quality of tehineh, a paste of dehulled roasted sesame seeds, from Saudi Arabia and other countries, were studied. Results showed 24.7% protein, 58.9% fat, 2.3% fiber, 3.0% ash and <1.0% moisture. The tehineh contained relatively high amounts (mg/100 g) of P (692), Mg (362), Fe (7.19), Cu (1.96), Mn (1.46) and Zn (7.82) and low amounts of Ca (61) and Se (0.05). Gas-liquid chromatography (GLC) analysis of the oil revealed percentages of 42.4 for oleic, 39.7 for linoleic, 9.8 for palmitic and 6.4 for stearic acid. Lysine was the only limiting amino acid with a chemical score of 64, while the sulphur-containing amino acids (methionine + cystine) and tryptophan were present in amounts exceeding the requirement of the FAO/WHO (1973) reference protein. The in vitro protein digestibility (IVPD) value of 83.3% and the calculated protein efficiency ratio (C-PER), 2.14, were slightly higher than those of sesame seeds and both values were lower than the IVPD and C-PER of 90.0% and 2.50, respectively, for ANRC casein.     

Nitrogen management in reservoir catchments through constructed wetland systems.

In this study, nitrogen removal was investigated in pilot-scale subsurface flow (SSF) and in free water surface flow (FWS) constructed wetlands installed in the campus of TUBITAK-Marmara Research Center, Gebze, near Istanbul, Turkey. The main purposes of this study are to apply constructed wetlands for the protection of water reservoirs and to reuse wastewater. Experiments were carried out at continuous flow reactors. The effects of the type of plants on the removal were investigated by using emergent (Canna, Cyperus, Typhia spp., Phragmites spp., Juncus, Poaceae, Paspalum and Iris.), submerged (Elodea, Egeria) and floating (Pistia, Salvina and Lemna) marsh plants at different conditions. During the study period HLRs were 30, 50, 70, 80 and 120 L m(2)d(-1) respectively. The average annual NH4-N, NO(3)-N, organic N and TN treatment efficiencies in SSF and FWS wetlands are 81% and 68%, 37% and 49%, 75% and 68%, 47% and 53%, respectively. Nitrification, denitrification and ammonification rate constant (k20) values in SSF and FNS systems have been found as 0.898 d(-1) and 0.541 d(-1), 0.488 d(-1) and 0.502 d(-1), 0.986 d(-1) and 0.908 respectively. Two types of the models (first-order plug flow and multiple regression) were tried to estimate the system performances.     

Hyperspectral imaging-based unsupervised adulterated red chili content transformation for classification: Identification of red chili adulterants

Neural Computing and Applications - Preserving red-chili quality is of utmost importance in which the authorities demand quality techniques to detect, classify, and prevent it from impurities. For...     

A cross-layer approach for partition detection at overlay layer for structured P2P in MANETs

Due to limited radio range and mobility of nodes in mobile ad hoc networks (MANETs), the network partitioning and merging could occur frequently. When structured peer-to-peer (P2P) overlays are running over MANETs, then network partition in the physical network can also cause network partition at the overlay layer. Existing approaches for structured P2P overlay over MANETs do not detect network partition at the overlay layer. This paper proposes a cross-layer approach to detect network partition at the overlay layer for structured P2P overlay over MANETs. Simulation results show that the proposed approach is highly effective and efficient in terms of routing overhead, success ratio and false-negative ratio.     

Leveraging Transfer Learning for Spatio-Temporal Human Activity Recognition from Video Sequences

Human Activity Recognition (HAR) is an active research area due to its applications in pervasive computing, human-computer interaction, artificial intelligence, health care, and social sciences. Moreover, dynamic environments and anthropometric differences between individuals make it harder to recognize actions. This study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world applications. It uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural network. Moreover, the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification information. Six state-of-the-art pre-trained models are exploited to find the best model for spatial feature extraction. For temporal sequence, this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short Term Memory (BiLSTM) to capture long-term dependencies. Two state-of-the-art datasets, UCF101 and HMDB51, are used for evaluation purposes. In addition, seven state-of-the-art optimizers are used to fine-tune the proposed network parameters. Furthermore, this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network (CNN), where two streams use RGB data. In contrast, the other uses optical flow images. Finally, the proposed ensemble approach using max hard voting outperforms state-of-the-art methods with 96.30% and 90.07% accuracies on the UCF101 and HMDB51 datasets.     

A model to measure QoE for virtual personal assistant

Until now the virtual assistants (like Siri, Google Now and Cortana) have primarily been confined to voice input and output only. Is there a justification for voice only confinement or can we enhance the user experience by adding a visual output? We hypothesized that providing a higher level of visual/auditory immersion would enhance the quality of user experience. In order to test this hypothesis, we first developed 4 variants of virtual assistant, each with a different audio/visual level of immersion. Developed virtual assistant systems were the following; audio only, audio and 2D visual display, audio and 3D visual display and audio and immersive 3D visual display. We developed a plan for usability testing of all 4 variants. The usability testing was conducted with 30 subjects against eight (8) dependent variables included presence, involvement, attention, reliability, dependency, easiness, satisfaction and expectations. Each subject rated these dependent variables based on a scale of 1–5, 5 being the highest value. The raw data collected from usability testing was then analyzed through several tools in order to determine the factors contributing towards the quality of experience for each of the 4 variants. The significant factors were then used develop a model that measures the quality of user experience. It was found that each variant had a different set of significant variables. Hence, in order to rate each system there is a need to develop a scale that is dependent upon the unique set of variables for the respective variant. Furthermore, it was found that variant 4 scored the highest rate for Quality of Experience (QoE). Lastly several other qualitative conclusions were also drawn from this research that will guide future work in the field of virtual assistants.     

On-line neuro-expert monitoring system for Borssele Nuclear Power Plant

A new method for an on-line monitoring system for the nuclear power plants has been developed utilizing the neural networks and the expert system. The integration of them is expected to enhance a substantial potential of the functionality as operators support.

The recurrent neural network and the feed-forward neural network with adaptive learning are selected for the plant modeling and anomaly detection because of the high capability of modeling for dynamic behavior. The expert system is used as a decision agent, which works on the information space of both the neural networks and the human operators. The information of other sensory signals is also fed to the expert system, together with the outputs that the neural networks generate from the measured plant signals. The expert system can treat almost all known correlation between plant status patterns and operation modes as a priori set of rules.

From the off-line test at Borssele Nuclear Power Plant (PWR 480 MWe) in the Netherlands, it was shown that the neuro-expert system successfully monitored the plant status. The expert system worked satisfactorily in diagnosing the system status by using the outputs of the neural networks and a priori knowledge base from the PWR simulator. The electric power coefficient is simultaneously monitored from the measured reactive and active electric power signals.     

Improvement of Grain Size and Dome Height of Microalloyed Steels Using Taguchi Method Based on Grey Relational Grade in Controlled Rolling

An efficient approach was introduced for improving the condition of major controlled rolling process parameters of roughing, finishing and coiling temperatures and optimizing these parameters to obtain minimum grain size and maximum dome height simultaneously. Taguchi method combined with grey relational analysis was applied to achieve optimum grain size and dome height during controlled rolling process. For this purpose, four levels for the above temperatures were chosen and sixteen experiments were conducted based on orthogonal array of Taguchi method. Based on Taguchi approach, signal-to-noise (S/N) ratios were calculated and used in order to obtain the optimum levels for every input parameter. Analysis of variance revealed that finishing and coiling temperatures have the maximum effect on the grain size and dome height of microalloyed steels. The confirmation tests with the optimal levels of parameters indicated that the grain size and dome height of controlled rolled microalloyed steels can be improved effectively through this approach.     

Solubility Enhancement of Fe in ZnO Nanoparticles Prepared by Co-Precipitation Method

Journal of Superconductivity and Novel Magnetism - Crystalline ZnO offers an excellent host matrix to create a dilute magnetic semiconductor (DMS) owing to its facile Zn-atom substitution with the...     

Estimation of bitumen emulsion dispersion using a light scattering method

Petroleum bitumen in the emulsified state is widely used as a binder in road and civil construction. Along with traditional quality parameters of bitumen emulsions the particle size is considered to be very important characteristic. The particle size of the dispersed phase affects the setting rate and viscosity of bitumen emulsions. The complete data on the particles size distribution can be obtained using such technical methods as light scattering and microscopy. In the study bitumen emulsions with three types of surfactants (cationic, nonionic, zwitterionic) have been prepared and their viscosity and decay index have been determined. The emulsions also have been studied with a microscope, as well as by the method of dynamic light scattering. The dynamics of particle size changes of the bitumen emulsions has been compared with the dynamics of the change in the basic parameters of the emulsions, such as viscosity and decay index, and the correlation between them has been found. Thus, using a modern laser and optical equipment, it is possible to estimate the quality of emulsions and, if necessary, to correct their composition.