Detection of edges from projections

In a number of applications of computerized tomography, the ultimate goal is to detect and characterize objects within a cross section. Detection of edges of different contrast regions yields the required information. The problem of detecting edges from projection data is addressed. It is shown that the class of linear edge detection operators used on images can be used for detection of edges directly from projection data. This not only reduces the computational burden but also avoids the difficulties of postprocessing a reconstructed image. This is accomplished by a convolution backprojection operation. For example, with the Marr-Hildreth edge detection operator, the filtering function that is to be used on the projection data is the Radon transform of the Laplacian of the 2-D Gaussian function which is combined with the reconstruction filter. Simulation results showing the efficacy of the proposed method and a comparison with edges detected from the reconstructed image are presented

Minimal sets of generators for the relation ideals of certain monomial curves

Let O be a monomial curve in the affine algebraic e-space over a field K and P be the relation ideal of O. If O is defined by a sequence of e positive integers some e - 1 of which form an arithmetic sequence then we construct a minimal set of generators for P and write an explicit formula for mu(P).

Reversible Insertion of a Trivalent Cation onto MnO2 Leading to Enhanced Capacitance

Supercapacitor properties of MnO2 are studied generally in aqueous alkali metal salt solutions, often in a Na salt solution. During electrochemical discharge-charge processes, Na+ ions from the electrolyte get reversibly inserted/deinserted on the surface of MnO2 particles, which leads to redox reaction between MnOONa and MnO2. In the present study, it has been shown that MnO2 exhibits enhanced capacitance behaviour in a rare earth metal salt solution, namely, La(NO3)(3) solution in comparison with NaNO3 and Mg(NO3)(2) aqueous solutions. The specific capacitance increases with an increase in charge on the solution cation (Na+, Mg2+ and La3+). It is proposed that the number of surface sites for adsorption of cations remains unaltered in all solutions. The surface insertion of cation facilitates the reduction of Mn4+ in MnO2 to Mn3+ equivalent to the charge present on the cation. As the specific capacitance is related to the quantity of charge that is exchanged between the solid MnO2 and the aqueous solution, the trivalent cation (La3+) provides greater specific capacitance than in Mg(NO3)(2) and NaNO3 electrolytes. Accordingly, the number of Mn(IV)/Mn(III) redox pairs involved in the neighbourhood of the adsorption site is one, two and three when Na+, Mg2+ and La3+ ions, respectively, are adsorbed. (C) 2011 The Electrochemical Society. DOI: 10.1149/1.3565177] All rights reserved.

Image reconstruction from incomplete projections data: A multiresolution modeling approach

Computerized tomography is an imaging technique which produces cross sectional map of an object from its line integrals. Image reconstruction algorithms require collection of line integrals covering the whole measurement range. However, in many practical situations part of projection data is inaccurately measured or not measured at all. In such incomplete projection data situations, conventional image reconstruction algorithms like the convolution back projection algorithm (CBP) and the Fourier reconstruction algorithm, assuming the projection data to be complete, produce degraded images. In this paper, a multiresolution multiscale modeling using the wavelet transform coefficients of projections is proposed for projection completion. The missing coefficients are then predicted based on these models at each scale followed by inverse wavelet transform to obtain the estimated projection data.

Studies on adsorption of guar gum onto biotite mica

The interaction of guar gum with biotite mica has been investigated through adsorption, flotation and electrokinetic measurements. The adsorption densities of guar gum increase with increase of pH and the isotherms exhibit Langmuirian behaviour. Pretreatment of mica with a complexing agent such as EDTA results in a decrease in the adsorption density, highlighting the contribution of metal ions to the adsorption process. An increase in the surface face-to-edge ratio lends to an increase in the adsorption density. The flotation recoveries decrease as a function of pH, complementing the adsorption results. However, polymer depressant ability is reduced in the case of EDTA treated mica, consequent to reduction of metallic sites. Electrokinetic measurements portray conformational rearrangements of macromolecules with the loading, resulting in the shift of the shear plane, further away from the interface. Dissolution experiments indicate release of metal ions from mica, while co-precipitation tests confirm polymer-metal ion interaction in the bulk solution. The adsorption process is governed by hydrogen bonding as well as chemical interaction between guar gum and the surface metal hydroxide groups of mica. (C) 1997 Published by Elsevier Science Ltd.

Mapping adaptive resonance theory onto ring and mesh architectures

In recent years, parallel computers have been attracting attention for simulating artificial neural networks (ANN). This is due to the inherent parallelism in ANN. This work is aimed at studying ways of parallelizing adaptive resonance theory (ART), a popular neural network algorithm. The core computations of ART are separated and different strategies of parallelizing ART are discussed. We present mapping strategies for ART 2-A neural network onto ring and mesh architectures. The required parallel architecture is simulated using a parallel architectural simulator, PROTEUS and parallel programs are written using a superset of C for the algorithms presented. A simulation-based scalability study of the algorithm-architecture match is carried out. The various overheads are identified in order to suggest ways of improving the performance. Our main objective is to find out the performance of the ART2-A network on different parallel architectures. (C) 1999 Elsevier Science B.V. All rights reserved.

Solution grafting of maleic anhydride onto polyethylene and compatibilization of polyetylene-nylon blends

Maleic anhydride (MAH) has been grafted onto high density polyethylene (HDPE) with benzoyl peroxide (BOP) initiator in toluene solution. Maximum degree of grafting (12%) without crosslinking has been obtained using MAH/HDPE and BOP/HDPE weight ratios of 1.0 and 0.15 respectively, at 110 degrees C. The HDPE-g-MAH compatibilizer is found to drastically reduce the dispersed phase size and also to produce homogeneous blends for relatively low concentrations of dispersed phase in HDPE/nylon blends. Addition of this compatibilizer results in increase of tensile strength and modulus with increasing nylon content of HDPE/nylon blends, while the opposite is found for the blends without any added compatibilizer.

Surface chemical studies on the competitive adsorption of poly(acrylic acid) and poly(vinyl alcohol) onto alumina

The adsorption of poly(acrylic acid) (PAA) and poly(vinyl alcohol) (PVA) onto alumina has been studied as a function of pH, both individually and in the presence of each other. The adsorption density of PAA is found to decrease with an increase of pH while that of PVA shows the opposite trend. In a binary system containing PAA and PVA, the presence of PVA does not affect the adsorption of PAA onto alumina, but the addition of PAA diminishes the adsorption of PVA in the pH range investigated. The adsorption isotherm of PAA at acidic pH exhibits high-affinity Langmuirian behavior. The isotherms for PVA appear rounded and are of the low-affinity type, Once again the adsorption isotherms of PAA remain unaltered in the presence of PVA whereas those of PVA are significantly affected resulting in a lowering of the adsorption density consequent to PAA addition. A variation in the sequence of addition of PAA and PVA does not affect the adsorption behavior of either of the polymers, The electrokinetic behavior of alumina with PAA is hardly influenced by the addition of PVA, On the other hand, the electrophoretic mobility of alumina in the presence of PVA is significantly altered in the presence of PAA and closely resembles the trend observed with PAA alone. Desorption studies reveal that over 80% of PVA could be desorbed in the pH range 3-9 whereas in the case of PAA, the percent desorption increases from 20 to about 70% as the pH is increased from about 3 to 8. Solution conductivity tests confirm interaction of aluminum species and PAA in the bulk solution. FTIR spectroscopic data provide evidence in support of hydrogen bonding and chemical interaction in the case of the PAA-alumina system and hydrogen bonding with respect to the PVA-alumina interaction. (C) 1999 Academic Press.

Efficient algorithms for learning kernels from multiple similarity matrices with general convex loss functions

In this paper we consider the problem of learning an n × n kernel matrix from m(1) similarity matrices under general convex loss. Past research have extensively studied the m = 1 case and have derived several algorithms which require sophisticated techniques like ACCP, SOCP, etc. The existing algorithms do not apply if one uses arbitrary losses and often can not handle m > 1 case. We present several provably convergent iterative algorithms, where each iteration requires either an SVM or a Multiple Kernel Learning (MKL) solver for m > 1 case. One of the major contributions of the paper is to extend the well knownMirror Descent(MD) framework to handle Cartesian product of psd matrices. This novel extension leads to an algorithm, called EMKL, which solves the problem in O(m2 log n 2) iterations; in each iteration one solves an MKL involving m kernels and m eigen-decomposition of n × n matrices. By suitably defining a restriction on the objective function, a faster version of EMKL is proposed, called REKL,which avoids the eigen-decomposition. An alternative to both EMKL and REKL is also suggested which requires only an SVMsolver. Experimental results on real world protein data set involving several similarity matrices illustrate the efficacy of the proposed algorithms.

Removal of As(V) by adsorption onto mixed rare earth oxides

Arsenic pollution of water is a major problem faced worldwide. Arsenic is a suspected carcinogen in human beings and is harmful to other living beings. In the present study, a novel adsorbent was used to remove arsenate [As(V)] from synthetic solutions. The adsorbent, which is a mixture of rare earth oxides, was found to adsorb As(V) rapidly and effectively. The effect of various parameters such as contact time, initial concentration, pH, and adsorbent dose on adsorption efficiency was investigated. More than 90% of the adsorption occurred within the first 10 min and the kinetic rate constant was found to be about 3.5 mg min(-1). Adsorption efficiency was found to be dependent on the initial As(V) concentration, and the adsorption behavior followed the Langmuir adsorption model. The optimum pH was found to be 6.5. The presence of other ions such as nitrate, phosphate, sulphate, and silicate decreased the adsorption of As(V) by about 20-30%. The adsorbed As(V) could be desorbed easily by washing the adsorbent with pH 12 solution. This study demonstrates the applicability of naturally occurring rare earth oxides as selective adsorbents for As(V) from solutions.

Grafting of m-isopropenyl-alpha, alpha-dimethylbenzyl-isocyanate (m-TMI) onto isotactic polypropylene: Synthesis and characterization

A novel vinyl monomer with an isocyanate functional group, m-isopropenyl-alpha,alpha-dimethylbenzyl-isocyanate (m-TMI), was grafted onto isotactic polypropylene (i-PP) using dicumyl peroxide (DCP) as the initiator. This would open up the possibility of using the grafted polymer with the reactive isocyanate group as compatibilizer for blending carbohydrates such as cellulose with. polypropylene. The grafting was carried out in a Brabender Plasticoder at 180degreesC. The effects of monomer and initiator concentrations on the yield of grafting were investigated by performing statistical analysis. While the grafting yield increased with the concentration of DCP at any given concentration of m-TMI, the variation of the grafting yield with m-TMI concentration, for a given concentration of DCP, went through a maximum, the optimum yield of 7.8% (w/w) being obtained at 10 wt.% concentration of both DCP and m-TMI. The grafting reaction is. accompanied by considerable chain scission of I-PP, resulting in a decrease in the molecular weight of the grafted polymer. While the molecular weight drops sharply even at a low concentration of DCP, there occurs no further significant change in the molecular weight even at much higher concentrations of the initiator.

Tree structure for efficient data mining using rough sets

In data mining, an important goal is to generate an abstraction of the data. Such an abstraction helps in reducing the space and search time requirements of the overall decision making process. Further, it is important that the abstraction is generated from the data with a small number of disk scans. We propose a novel data structure, pattern count tree (PC-tree), that can be built by scanning the database only once. PC-tree is a minimal size complete representation of the data and it can be used to represent dynamic databases with the help of knowledge that is either static or changing. We show that further compactness can be achieved by constructing the PC-tree on segmented patterns. We exploit the flexibility offered by rough sets to realize a rough PC-tree and use it for efficient and effective rough classification. To be consistent with the sizes of the branches of the PC-tree, we use upper and lower approximations of feature sets in a manner different from the conventional rough set theory. We conducted experiments using the proposed classification scheme on a large-scale hand-written digit data set. We use the experimental results to establish the efficacy of the proposed approach. (C) 2002 Elsevier Science B.V. All rights reserved.

Dynamic contact angle beating from drops impacting onto solid surfaces exhibiting anisotropic wetting

This paper reports an experimental investigation of low Weber number water drops impacting onto solid surfaces exhibiting anisotropic wetting. The wetting anisotropy is created by patterning the solid surfaces with unidirectional parallel grooves. Temporal measurements of impacting drop parameters such as drop base contact diameter, apparent contact angle of drop, and drop height at the center are obtained from high-speed video recordings of drop impacts. The study shows that the impact of low Weber number water drops on the grooved surface exhibits beating phenomenon in the temporal variations of the dynamic contact angle anisotropy and drop height at the center of the impacting drop. It is observed that the beating phenomenon of impacting drop parameters is caused by the frequency difference between the dynamic contact angle oscillations of impacting drop liquid oriented perpendicular and parallel to the direction of grooves on the grooved surface. The primary trigger for the phenomenon is the existence of non-axisymmetric drop flow on the grooved surface featuring pinned and free motions of drop liquid in the directions perpendicular and parallel to the grooves, respectively. The beat frequency is almost independent of the impact drop Weber number. Further experimental measurements with solid surfaces of different groove textures show that the grooved surface with larger wetting anisotropy may be expected to show a dominant beating phenomenon. The phenomenon is gradually damped out with time and is fully unrecognizable at higher drop impact Weber numbers. (C) 2011 Elsevier B.V. All rights reserved.

A Randomized Algorithm for Large Scale Support Vector Learning

We propose a randomized algorithm for large scale SVM learning which solves the problem by iterating over random subsets of the data. Crucial to the algorithm for scalability is the size of the subsets chosen. In the context of text classification we show that, by using ideas from random projections, a sample size of O(log n) can be used to obtain a solution which is close to the optimal with a high probability. Experiments done on synthetic and real life data sets demonstrate that the algorithm scales up SVM learners, without loss in accuracy. 1

A Fast Linear Separability Test by Projection of Positive Points on Subspaces

A geometric and non parametric procedure for testing if two finite set of points are linearly separable is proposed. The Linear Separability Test is equivalent to a test that determines if a strictly positive point h > 0 exists in the range of a matrix A (related to the points in the two finite sets). The algorithm proposed in the paper iteratively checks if a strictly positive point exists in a subspace by projecting a strictly positive vector with equal co-ordinates (p), on the subspace. At the end of each iteration, the subspace is reduced to a lower dimensional subspace. The test is completed within r ≤ min(n, d + 1) steps, for both linearly separable and non separable problems (r is the rank of A, n is the number of points and d is the dimension of the space containing the points). The worst case time complexity of the algorithm is O(nr3) and space complexity of the algorithm is O(nd). A small review of some of the prominent algorithms and their time complexities is included. The worst case computational complexity of our algorithm is lower than the worst case computational complexity of Simplex, Perceptron, Support Vector Machine and Convex Hull Algorithms, if d