Emotion Recognition from Occluded Facial Images Using Deep EnsembleModel

Facial expression recognition has been a hot topic for decades, but high intraclass variation makes it challenging. To overcome intraclass variation for visual recognition, we introduce a novel fusion methodology, in which the proposed model first extract features followed by feature fusion. Specifically, RestNet-50, VGG-19, and Inception-V3 is used to ensure feature learning followed by feature fusion. Finally, the three feature extraction models are utilized using Ensemble Learning techniques for final expression classification. The representation learnt by the proposed methodology is robust to occlusions and pose variations and offers promising accuracy. To evaluate the efficiency of the proposed model, we use two wild benchmark datasets Real-world Affective Faces Database (RAF-DB) and AffectNet for facial expression recognition. The proposed model classifies the emotions into seven different categories namely: happiness, anger, fear, disgust, sadness, surprise, and neutral. Furthermore, the performance of the proposed model is also compared with other algorithms focusing on the analysis of computational cost, convergence and accuracy based on a standard problem specific to classification applications.     

An optimal integrated lot sizing and maintenance strategy for multi-machines system with energy consumption

This paper proposes an integrated model for multi-machines dynamic lot sizing aiming to produce a single item, considering the energy consumption during the production horizon. The objective is to find, firstly, the optimal lot size as well as the number of machines that satisfy a random demand under given service level and secondly, maintenance plan depended to production planning to minimise the total production, energy and maintenance costs. In fact, the problem of energy consumption is one of the most evoked topics especially with the decision of many governments to reduce theirs (For example France is willing to reduce the total consumption by 20% by 2020). The keys of this study are to consider, firstly, the correlation between the forecasting of demand, the variation of the working machines as well as their production rates under energy constraint and secondly the correlation between the production cadences and the maintenance strategy of all machines.     

Quality issue in forecasting problem of production and maintenance policy for production unit

In this study, we consider an unreliable deteriorating production system that produces conforming and non-conforming products to satisfy a random demand under a given service level and during a finite horizon. The production system is subjected to a failure-prone machine. The quality of the produced products is affected by the machine deterioration since the rate of defectives increases as the deterioration increases. Preventive maintenance actions can be piloted on the production system to reduce the influence of deterioration and the defective rate. A joint control policy is based on a stochastic production and maintenance planning problem with goals to determine, firstly, the economic plan of production and secondly, the optimal maintenance strategy. The proposed jointly optimisation minimises the total cost of production, inventory, maintenance and defectives. A failure rate and quality relationship are defined to show the influence of the production rates variation on the failures rate as well as on the defective rate. A numerical example and an industrial case study are adopted to illustrate the proposed approach and a sensitivity analysis to validate the jointly optimisation.     

The transverse elastic properties of chiral honeycombs

This work describes the out-of-plane linear elastic mechanical properties of trichiral, tetrachiral and hexachiral honeycomb configurations. Analytical models are developed to calculate the transverse Young’s modulus and the Voigt and Reuss bounds for the transverse shear stiffness. Finite Element models are developed to validate the analytical results, and to identify the dependence of the transverse shear stiffness vs. the gauge thickness of the honeycombs. The models are then validated with experimental results from flatwise compressive and simple shear tests on samples produced with rapid prototype (RP)-based techniques.     

Microwave synthesis of large few-layer graphene sheets in aqueous solution of ammonia

Few-layer graphene (FLG) sheets with sizes exceeding several micrometers have been synthesized by exfoliation of expanded graphite in aqueous solution of ammonia under microwave irradiation, with an overall yield approaching 8 wt.%. Transmission electron microscopy (in bright-field and dark-field modes) together with electron diffraction patterns and atomic force microscopy confirmed that this graphene material consisted mostly of mono-, bi- or few-layer graphene (less than ten layers). The high degree of surface reduction was confirmed by X-ray photoelectron and infrared spectroscopies. In addition, the high stability of the FLG in the liquid medium facilitates the deposition of the graphene material onto several substrates via low-cost solution-phase processing techniques, opening the way to subsequent applications of the material.      

Hepatitis C virus RNA in serum of blood donors with or without elevated transaminase levels

The pattern of hepatitis C virus (HCV) viremia in blood donors who are positive for antibody to HCV (anti-HCV) according to the level of transaminase activity is unclear. A polymerase chain reaction-based HCV RNA detection method was used to study two clearly defined groups of anti-HCV-positive blood donors with repeatedly normal (n = 27) and elevated (n = 17) alanine aminotransferase (ALT) levels. HCV RNA was detected in only 4 of 27 blood donors with normal ALT values and 15 of 17 with elevated ALT values. These results indicate that anti-HCV-positive blood donors with normal ALT levels constitute a heterogeneous group, as HCV viremia is detectable in only a small proportion of cases. Polymerase chain reaction should be useful in the surveillance of anti-HCV-positive blood donors with normal ALT levels, by identifying those who might benefit from further investigation and treatment.     

Purification,structural data and biological properties of polysaccharide from Prunus amygdalus gum

This work demonstrates the efficiency of almond gum polysaccharides (AGPs) as bioactive compounds. AGPs were first extracted using H₂O₂, in the presence of NaOH, at different times and temperatures. The optimal extraction conditions were 4% H₂O₂ and 2 N NaOH, for 7 h at 50 °C, leading to an extraction yield of 58.2% (w/w). After a purification step, the retained AGPs were characterised using high‐performance liquid chromatography showing a molecular weight of 99.3 kDa. The monosaccharide composition of AGPs were assessed using gas chromatography–mass spectrometry. AGPs were found to be a complex heteropolysaccharide with a repeating unit mainly composed of galactose, arabinose, xylose, mannose, rhamnose, and glucuronic acid with the respective ratios: 45:26:7:10:1:11. The acidic nature of the polysaccharide is due to the presence of glucuronic acid. Total antioxidant activity, free radical‐scavenging activity and reducing power assay of AGPs were investigated. The obtained results showed high antioxidant activities of AGPs. Furthermore, beyond 60 mg mL^?1, AGPs exhibited bacterial growth inhibition for five pathogenic strains: Escherichia coli, Staphylococcus aureus, Enterococcus feacalis, Pseudomonas aeruginosa and Salmonella typhimurium.     

Enthalpies of formation of liquid and solid (gallium + palladium) alloys

The enthalpies of formation of liquid (Ga + Pd) alloys were determined by direct reaction calorimetry in the temperature range 1322 <T/K < 1761 and the molar fraction range 0 <x _Pd < 0.87. The enthalpies are very negative with a minimum Δ_mix H _m = −70.4 ± 3.0 kJ mol_-1 atx _Pd = 0.6, independent of the temperature. Limiting partial molar enthalpies of palladium and gallium were calculated as Δh _m (Ga liquid in ∞liquid Pd) = −265 ± 10 kJ mol₋₁ and Δh _m (Pd liquid in ∞liquid Ga) = -144 ± 5 kJ mol₋₁. The integral molar enthalpy is given by Δ_mix H _m =x(1-x) (-143.73 -232.47x + 985.77x ²-4457.8.x ³ + 6161.1x ⁴ + 2577.4x ⁵), wherex = x _Pd. Moreover, values for the enthalpies of formation and fusion of PdGa, Pd₂Ga, and the solid solution (withx _Pd = 0.8571) have been proposed. These results have been discussed taking into account the equilibrium phase diagram. Formerly Ph.D. student, Université de Provence