Two-tier architecture for unconstrained handwritten character recognition

In this paper, we propose an approach that combines the unsupervised and supervised learning techniques for unconstrained handwritten numeral recognition. This approach uses the Kohonen self-organizing neural network for data classification in the first stage and the learning vector quantization (LVQ) model in the second stage to improve classification accuracy. The combined architecture performs better than the Kohonen self-organizing map alone. In the proposed approach, the collection of centroids at different phases of training plays a vital role in the performance of the recognition system. Four experiments have been conducted and experimental results show that the collection of centroids in the middle of the training gives high performance in terms of speed and accuracy. The systems developed also resolve the confusion between handwritten numerals.     

Development of nano indium tin oxide (ITO) grains by alkaline hydrolysis of In(III) and Sn(IV) salts

Indium tin oxide (ITO) nano powders of different compositions (In: Sn = 90: 10, 70: 30 and 50: 50) were prepared by heat treatment (300-450°C) of mixed hydroxides of In(III) and Sn(IV). The hydroxides were obtained by the reaction of aq. NH₃ with mixed aq. solutions of In(NO₃)₃ and SnCl₄. FTIR and TG/DTA studies revealed that powders existed as In(OH)₃H₂O—SnO₃H₂H₂O in the solid state and then they transformed to In₂O₃—SnO₂ via some metastable intermediates after 300°C. Cubic phase of In₂O₃ was identified by XRD for the oxides up to 30% of Sn. Particle size measurements of the solid dispersed in acetone and SEM study for microstructure showed that the oxides were in the nano range (55-75 nm) whereas the size range determined from Debye-Scherrer equation were 11–24 nm.     

A computational study of LP-based heuristic algorithms for two-dimensional guillotine cutting stock problems

In this paper we develop and compare several heuristic methods for solving the general two-dimensional cutting stock problem. We follow the Gilmore-Gomory column generation scheme in which at each iteration a new cutting pattern is obtained as the solution of a subproblem on one stock sheet. For solving this subproblem, in addition to classical dynamic programming, we have developed three heuristic procedures of increasing complexity, based on GRASP and Tabu Search techniques, producing solutions differing in quality and in time requirements. In order to obtain integer solutions from the fractional solutions of the Gilmore-Gomory process, we compare three rounding procedures, rounding up, truncated branch and bound and the solution of a residual problem. We have coded and tested all the combinations of algorithms and rounding procedures. The computational results obtained on a set of randomly generated test problems show their relative efficiency and allow the potential user to choose from among them, according to the available computing time. Rceived: January 9, 2001 / Accepted: December 10, 2001     

Stress resultant geometrically non-linear shell theory with drilling rotations. Part III: Linearized kinematics

A consistent formulation of the geometrically linear shell theory with drilling rotations is obtained by the consistent linearization of the geometrically non-linear shell theory considered in Parts I and II of this work. It was also shown that the same formulation can be recovered by linearizing the governing variational principle for the three-dimensional geometrically non-linear continuum with independent rotation field. In the finite element implementation of the presented shell theory, relying on the modified method of incompatible modes, we were able to construct a four-node shell element which delivers a very high-level performance. In order to simplify finite element implementation, a shallow reference configuration is assumed over each shell finite element. This approach does not impair the element performance for the present four-node element. The results obtained herein match those obtained with the state-of-the-art implementations based on the classical shell theory, over the complete set of standard benchmark problems.     

Mathematical analysis of viscometric (polymer) flow fields in capillaries: Taylor dispersion revisited

A general formulation for capillary flow of two miscible fluids – one a dilute plug of polymer fluid inserted into a fully developed Poiseuille flow of the other, a Newtonian stream – is examined for its long time behavior. Phenomenologically, the system evolves from an initial state, that of a plug within the boundaries of sharp, well defined fronts inside a Newtonian stream, to a more homogenized state in the very long time scale. This problem was addressed by G.I. Taylor but with regard to a system of two Newtonian fluids, leading to the well-known results commonly described as `Taylor axial dispersion'. In this paper, a general and systematic perturbation analysis is presented from which Taylor's result is recovered as a special case of a more general solution which applies to fluids incorporating elastic properties. In particular, the influence of viscoelasticity and (polymer) diffusivity on the observed pressure profile in the capillary conduit is examined. This effect is clearly separated out for small Peclet number flows using asymptotic and numerical analysis. The results identify the influence of fluid viscosity, elasticity, and diffusivity on the observed pressure profile and form the basis for the improved characterization of polymeric elasticity using capillaries – a finding that is of significant scientific and commercial interest. These results were obtained by the authors as a class of observations resulting from the perturbation analysis of forced-flow capillary devices in viscoelastic fluid property investigation.     

Influence of fretting on the fatigue strength at the vise clamp-specimen interface

Fretting fatigue is one of the most important phenomena for inducing a significant reduction of fatigue strength and consequently, leading to unexpected failure accidents of the engineering structures even at very low stresses. In the present study, both plain and fretting fatigue tests with zero mean stress were carried out on two different types of steel, low-carbon steel and martensitic stainless steel, by means of a reversed bending fatigue testing machine. The drop in the fatigue strengths through fretting at vise clamp-specimen interface were significant for both tested steels. The fretting processes produced a reduction in fatigue strength of about 27% for low-carbon steel and 16% for martensitic stainless steel.