In this paper, a novel approach for fusing shape and texture Local Binary Pattern (LBP) for 3D Face Recognition is presented. Using the recently proposed mesh-LBP [1], it is now possible to compute LBP directly on a mesh manifold, allowing Features Early- Fusion to enhance face description power. Compared to its depth image counterpart, the proposed method is distinguished by a) inherits the intrinsic advantages of mesh surfaces, b) can accommodate partial or rotation matching, c) allows early-level fusion of texture and shape descriptors. An experimentation of early-fusion techniques over Bosphorus database is presented.
Smart grid is emerging as the future power grid, with smart techniques to optimize power consumption and electricity generation. Minimizing the peak power consumption under a fixed delay requirement is a significant problem in the smart grid. For this problem, all appliances must be scheduled within a given finite time duration. We consider the problem of minimizing the peak demand under appliances constraints by scheduling power jobs with uniform release dates and deadlines. As the problem is known to be NP-Hard, we propose two versions of a greedy approximation algorithm for solving this problem. Our theoretical analysis and experimental results show that our proposed heuristics outperform existing methods by providing a better approximation to the optimal solution.
Social influence and persuasive technology have been applied widely in commercial and healthcare promotions, but not so much for educational purposes. Social context is crucial in educational environments to define important properties of persuasion power that could lead to increasing retention of learning and behavioral skills. In this paper, we propose a Behavioral Change Support System for Career-Adoption (BCSS-CA) to encourage learners in higher education to adopt desired career behaviors according to current local market needs. We designed a social network analysis (SNA) method to extract the most influential nodes in a SN using different strategies; which then are used as an initial set for influence diffusion models to compare their social influence performance. The results revealed that nodes of highest "coreness" value have the highest impact in influencing other nodes under both models. These nodes in the are then "hired" as "persuasive agents" in our BCSS-CA.
Abstract?Detection of defective chips on wafers is a crucial issue to semiconductor manufacturing companies in order to track back the root of the failure. Different machine learning approaches have been applied for detecting defect patterns in semiconductor wafer maps. This paper applies a more advanced machine learning algorithm (deep learning) using a convolutional neural networks (CNN) to detect the defect patterns. Moreover, it considers more realistic setting by adding composite or mixed defect patterns to the dataset. CNN achieves accuracy of 85%, which confirms better performance than the fully connected neural networks, such as Multilayer perceptron (MLP) and Radial based function network (RBF).
Human detection within a scene can be implemented based on an algorithm by Peer et al [1]. This algorithm detects the skin color in RGB images by checking some pixel conditions, any pixel value satisfying those conditions will be labeled as skin. Mahalanobis distance [2]; on the other hand, is also used for skin detection. Mahalanobis distance depends on the mean (m) and covariance (c) of pixel values representing skin. This paper examine and compares these two methods in terms of methodology and qualitative assessment.
CD44, a transmembrane protein presented on the surface of many immune cells and cancer cells, plays a crucial role in cell-cell and cell-matrix interactions. The principal ligand for CD44 is the ubiquitous ECM glycosaminoglycan, Hyaluronic acid (HA). The interaction between CD44 and HA is implicated in various physiological and pathological processes including inflammation and cancer progression, respectively. The HA binding domain of CD44 is termed the link module, a domain that is conserved among HA-binding proteins. This paper investigated the conservation of this domain by conducting multiple sequence analysis across mammalian species where it was found that this domain is highly conserved as the similarity between the species exceeded 80%. Moreover, this study addressed the structural-functional analysis of HA-binding domain of human CD44 by predicting the potential disulfide bridges and N-linked glycosylation sites and relating it to the tertiary structure of HA-binding domain of CD44.
The CD24 is a small and highly glycosylated cell-surface protein that is linked via glycosyl-phosphatidylinositol (GPI) anchor. An interacting partner of CD24 is Lyn kinase; an intracellular Src-Family kinase, which is found to interact with the extracellular CD24 molecule via membrane rafts. Interaction between Lyn and CD24 in some human mutagenesis highlights the importance of analyzing Lyn structure. In this study, we have examined CD24, Lyn kinase and their possible interactions. Furthermore, conservational sequence analysis was computationally conducted for CD24 and Lyn kinase across species. The unavailability of CD24 structure in Protein Data Bank motivated the use of bioinformatics tools to predict CD24 structure. We performed visual analyses of Dasatinib-Lyn kinase complex structure, and proposed a suggestion for Dasatinib-based drug optimization. Understanding CD24 and Lyn kinase structures can assist in probing the functions and therapeutic strategies in cancer-linked pathways.
Seed germination was always a critical topic for propagating desert plants. Such seeds have limited access to fresh water resources and exposed to harsh environmental conditions (e.g. high temperatures), which resulted in developing many survival mechanisms; to adapt the surrounding stressful conditions and to naturally conserve species availability. Seed germination topic is of great interest, especially while dealing with medicinal crops, that have great economical values (e.g. pharmaceutical and aroma therapy). The main purpose of this work is to conduct seed germination studies on testing the effect of growing media and seed color on seedling emergence of Cleome brachycarpa, indigenous to the United Arab Emirates (UAE).
Epilepsy is a chronic disorder characterized by repetitive seizures. Available AEDs are effective in 60-80% of patients but associated with diverse side effects. Pitolisant, a H3R antagonist, was effective in different animal seizure models and in patients with photosensitive epilepsy. In the current study, pharmacophores of pitolisant were combined with that of safinamide or phenytoin with different spacer moieties. These novel ligands demonstrated good affinity and selectivity profiles. These agents were also investigated in different seizure models. Safinamide-related derivatives with bulky substituents significantly and dose-dependently reduced or fully protect from MES- and PTZ-induced convulsions with some in-vivo stereochemical preferences.
Microbially induced calcite precipitation has emerged as a soil strengthening technique with foreseen application to a vast range of geo-engineering problems. The technique introduces the bio-cemented geomaterial which is characterized by a peculiar structure due to the formation of CaCO3 mineral crystals among the soil grains. This study investigates the effect of the initial porosity and material intrinsic properties on the formation of the calcite solid matrix which governs the mechanical response. The same treatment conditions are applied herein to geomaterials of different grain size and porosity. The obtained bio-improved samples are subjected to drained triaxial shear and improved mechanical characteristics are obtained. Microstructural observations through scanning electron microscopy and energy dispersive X-ray analyses are carried out. Results allow obtaining a comprehensive understanding of the distinct geometrical and spatial characteristics of the newly formed structure and associate these characteristics with the obtained mechanical response and with the initial material properties.
