Ridge regression-based feature extraction for hyperspectral data

Feature extraction based on ridge regression (FERR) is proposed in this article. In FERR, a feature vector is defined in each spectral band using the mean of all classes in that dimension. Then, it is modelled using a linear combination of its farthest neighbours from among other defined feature vectors. The representation coefficients obtained by solving the ridge regression model compose the projection matrix for feature extraction. FERR can extract each desired number of features while the other methods such as linear discriminant analysis (LDA) and generalized discriminant analysis (GDA) have limitations in the number of extracted features. Experimental results on four popular real hyperspectral images show that the efficiency of FERR is superior to those of other supervised feature extraction methods in small sample-size situations. For example, for the Indian Pines dataset, the comparison between the highest average classification accuracies achieved by different feature extraction methods using a support vector machine (SVM) classifier and 16 training samples per class shows that FERR is 7% more accurate than nonparametric weighted feature extraction and is also 9% better than GDA. LDA, having the singularity problem in the small sample-size situations, has 40% less accuracy than FERR. The experiments show that generally the performance of FERR using the SVM classifier is better than when using the maximum likelihood classifier.     

A NOVEL EMBEDDED FEATURE SELECTION METHOD: A COMPARATIVE STUDY IN THE APPLICATION OF TEXT CATEGORIZATION

In text classification based on a vector space model, the high dimension of the feature may pose some problems. These problems occur not only for computational reasons, but also because of overfitting. Feature selection is an important preprocessing step used for text classification applications to reduce the vector space size, control the computational time, and maintain or improve performance. In this study, we used an embedded approach in feature selection in which the Chi-square (CHI) feature selector is a filter step. In this step, the less discriminative features are discarded. In the wrapper step, a novel algorithm is proposed based on the combination of the fast global search ability of the genetic algorithm (GA) and the positive feedback mechanism of ant colony optimization (ACO). In order to validate our approach, we carried out a series of experiments on Reuters-21578 corpus, and we compare the achieved results with some other well-known techniques. The evaluation results are such that our method obtained a better performance compared with the other methods in the majority of cases.     

High-dimensional image data feature extraction by double discriminant embedding

We propose a supervised feature extraction method in this paper that uses two successive transformations to produce the extracted features. The first projection maximizes the difference between spectral features. Thus, produced features have minimum overlap in the new feature space. The second projection maximizes the discrimination between classes. The proposed method, which is called double discriminant embedding (DDE), uses just the first statistics of data. Thus, DDE has good efficiency using limited training samples. The experimental results on four popular hyperspectral images show the better efficiency of DDE in comparison with LDA, GDA, NWFE, and supervised LPP methods in small sample size situation.     

High Performance Image Super-Resolution Using Convolutional Neural Networks and Nonsubsampled Contourlet Transform

Journal of Communications Technology and Electronics - Disadvantages of deep convolutional neural networks (CNNs) in image super-resolution (SR) are slow convergence rate, long training time, slow...     

A scoping review of the use of lab streaming layer framework in virtual and augmented reality research

Virtual Reality - The use of multimodal data allows excellent opportunities for human–computer interaction research and novel techniques regarding virtual and augmented reality (VR/AR)...     

Real-time P-H curve CNC interpolators for high speed cornering

In this paper, real-time Pythagorean hodograph (P-H) curve CNC interpolators are used for high speed corner machining. There are large contouring errors around sharp corners when low-bandwidth servo controllers (such as P-PI control) are used. In order to decrease the amount of the cornering errors, an improved interpolation method is proposed. The first deceleration phase of motion and the over-corner P-H curve constructed for regions with sharp corners are devised by quadratic and constant velocity interpolation algorithms, respectively. The geometric parameters of the over-corner P-H curve and the feed rate along the modified tool path are computed by pattern search algorithm in order to reduce the maximum cornering error. The proposed interpolation algorithm is implemented for symmetrical and unsymmetrical corners. The results of simulation, such as the cornering error and the total cornering time, are compared with previously published methods. It has been observed that the developed over-corner P-H approach can substantially reduce the amount of cornering error.     

Resource placement in Cartesian product of networks

The utilization of the limited resources of a multiprocessor or multicomputer system is a primary performance issue which is crucial for the design of many scheduling algorithms. While many of the existing parallel machines benefit from a regular product network topology, almost none of the previous resource placement techniques have come to recognize and exploit this inherent regularity. This paper introduces several novel algorithms for deriving resource placement schemes in product networks based on the assumption of perfect resource placement in their underling basic graphs. Our techniques use known schemes for the basic networks as their building blocks for deploying the resource placement scheme in the product network. This seriously cuts down the expenses required for deploying and rescaling the network. In particular, we propose some efficient algorithms for adjacency placement in a product of k$k$ heterogeneous graphs. Furthermore, we extend our approach and present algorithms for distant resource placement in product networks.     

Transplantation of nano-bioglass/gelatin scaffold in a non-autogenous setting for bone regeneration in a rabbit ulna

Bioactive glass has been investigated for variety of tissue engineering applications. In this study, fabrication, in vitro and in vivo evaluation of bioactive glass nanocomposite scaffold were investigated. The nanocomposite scaffolds with compositions based on gelatin and bioactive glass nanoparticles were prepared. The apatite formation at the surface of the nanocomposite samples confirmed by Fourier transform infrared spectroscopy, scanning electron microscopy and X-ray powder diffraction analyses. The in vitro characteristics of bioactive glass scaffold as well as the in vivo bone formation capacity of the bioactive glass scaffold in rabbit ulnar model were investigated. The bioactive glass scaffold showed no cytotoxicity effects in vitro. The nanocomposite scaffold made from gelatin and bioactive glass nanoparticles could be deliberated as an extremely bioactive and prospective bone tissue engineering implant. Bioactive glass scaffolds were capable of guiding bone formation in a rabbit ulnar critical-sized-defect model. Radiographic evaluation indicated that successful bridging of the critical-sized defect on the sides both next to and away from the radius took place using bioactive glass scaffolds. X-ray analysis also proposed that bioactive glass scaffolds supported normal bone formation via intramembranous formation     

Rheokinetics in curing process of polyfurfuryl alcohol: effect of homologous acid catalysts

Curing reaction of polyfurfuryl alcohol (PFA) resin was investigated using rheological measurements in the presence of different acid catalysts. A homologous series of dicarboxylic acid catalysts with even number of carbons of 2, 4 and 6 were chosen, i.e., oxalic acid (OX, 2 carbon), succinic acid (SU, 4 carbon) and adipic acid (AD, 6 carbon). Acidity of these catalysts in terms of pK _a was in the range of 1.25–4.43. Firstly, non-isothermal curing kinetics were investigated, and then, isothermal studies were performed at three different temperatures using 3 and 5 wt% of each acid catalyst. Isothermal cure behavior was best described by an empirical Arrhenius model of viscosity. An isoconversional method was applied to compute the changes in the effective activation energy as a function of degree of conversion. The change in activation energy at 50–10 kJ mol^?1 was observed for PFA/3 wt% OX system due to the vitrification and diffusion-controlled processes. Gel time, t _gel, was determined as a function of curing temperature, acid catalyst type and its concentration. The gelation time was consistent for all the formulations with an extent of conversion of about 0.7. Significant differences were observed in the curing behavior of resins in the presence of different acid catalysts. Increasing of the curing temperature significantly accelerated the curing process and increased curing rate constant of the PFA resin.     

Chondrogenic stimulation in mesenchymal stem cells using scaffold-based sustained release of platelet-rich plasma

Developing minimal invasive strategies via injectable hydrogels for effective repairing of cartilage defects is highly desired. Injectable hydrogels, which can simultaneously embed cell and growth factors (GFs), serve as in situ formed scaffolds and could support an accelerated tissue regeneration process. The purpose of this study is to fabricate a composite injectable hydrogel, based on alginate (Alg)/polyvinyl alcohol (PVA) incorporating platelet rich plasma (PRP)-encapsulated Alg sulfate (AlgS) microbeads, as a localized sustained release system of GFs, for the articular cartilage regeneration. The results show that synthesized AlgS microbeads support the sustained release of PRP GFs during 14 days, where preserve the bioactivity of them more than the free PRP. Rabbit adipose-derived mesenchymal stem cells in contact with PRP-loaded AlgS beads show more proliferation (2.7 folds) and have obviously higher deposition of collagen type ΙΙ and GAGs than free PRP treated ones. In addition, cells encapsulated into the hydrogel including PRP sustained release system show upregulated expression of collagen type ΙΙ (61 folds), Aggrecan (294 folds) and SOX9 (71.5 folds), as cartilage-critical genes, compared to the direct treatment by PRP. To summarize, the developed hybrid Alg/PVA hydrogel embedding with PRP-encapsulated AlgS microbeads is suggested as a potential in situ formed scaffold for cartilage regeneration.