GP-based secondary classifiers

Genetic programming (GP) is used to evolve secondary classifiers for disambiguating between pairs of handwritten digit images. The inherent property of feature selection accorded by GP is exploited to make sharper decision between conflicting classes. Classification can be done in several steps with an available feature set and a mixture of strategies. A two-step classification strategy is presented in this paper. After the first step of the classification using the full feature set, the high confidence recognition result will lead to an end of the recognition process. Otherwise a secondary classifier designed using a sub-set of the original feature set and the information available from the earlier classification step will help classify the input further. The feature selection mechanism employed by GP selects important features that provide maximum separability between classes under consideration. In this way, a sharper decision on fewer classes is obtained at the secondary classification stage. The full feature set is still available in both stages of classification to retain complete information. An intuitive motivation and detailed analysis using confusion matrices between digit classes is presented to describe how this strategy leads to improved recognition performance. In comparison with the existing methods, our method is aimed for increasing recognition accuracy and reliability. Results are reported for the BHA test-set and the NIST test-set of handwritten digits.     

Micropore volumes of carbons by means of a new isotherm equation

The linearized form of the isotherm equation developed by John is given by log log P = C + n log v. The log v values corresponding to relative pressure 1, on the extrapolated straight line of the experimental plot of the above equation gave micropore volume in good agreement with the values obtained by Dubinin method, especially for microporous carbons giving Type 1 and extremely low pressure isotherms. It is shown mathematically that the new equation is fundamentally the same as Dubin equation.     

Three-Dimensional Electromechanical Impedance Model for Multiple Piezoceramic Transducers—Structure Interaction

In the last few decades, piezoceramic (PZT) transducers have been extensively used either as actuators or as sensors in the vibration and noise control of aero, civil, and mechanical (ACM) systems. Only in the last decade, PZT transducers have been used in the electromechanical impedance (EMI) models as both sensor and actuator for the structural health monitoring of ACM systems. In the EMI models, the PZT transducers are generally surface bonded to the host structure and are then subjected to one-dimensional (1D) voltage to interrogate the structure for the desired frequency range. The interrogation results in the prediction of electromechanical admittance signatures. These signatures serve as indicators of the health/integrity of the structure. However, the existing single PZT–structure interaction models consider the PZT transducer to be negligible in mass and thus ignored. Moreover, they impose restrictions on the PZT shape, size, and isotropy. This paper presents a novel semianalytical multiple three-dimensional PZT–structure interaction model which considers the “mass” of the PZT transducers, and the transducers are subjected to parallel (1D) sinusoidal voltage. Further, the model does not impose restriction on the shape (square or rectangular), size (thin or thick), and electrical properties (isotropic or anisotropic) of PZT. The derived model is also experimentally verified using lab-sized aluminum plate. As it is generic, the model is expected to be applicable for the nondestructive evaluation of most ACM systems.     

Epitaxial integration of the highly spin-polarized ferromagnetic semiconductor EuO with silicon and GaN

Doped EuO is an attractive material for the fabrication of proof-of-concept spintronic devices. Yet for decades its use has been hindered by its instability in air and the difficulty of preparing and patterning high-quality thin films. Here, we establish EuO as the pre-eminent material for the direct integration of a carrier-concentration-matched half-metal with the long-spin-lifetime semiconductors silicon and GaN, using methods that transcend these difficulties. Andreev reflection measurements reveal that the spin polarization in doped epitaxial EuO films exceeds 90%, demonstrating that EuO is a half-metal even when highly doped. Furthermore, EuO is epitaxially integrated with silicon and GaN. These results demonstrate the high potential of EuO for spintronic devices.     

Effects of cooking methods on polyphenols,pigments and antioxidant activity in potato tubers

Potato tubers, which are one of the richest sources of antioxidants, are always cooked before human consumption. The objective of this study was to understand the effects of various domestic cooking methods, i.e., boiling, microwaving and baking on total phenolics, flavonoids, flavonols, lutein, anthocyanins and antioxidant activities in 5 cultivars and 9 advanced selections with different skin and flesh colors after 6 months of storage. The three cooking methods reduced the levels of these compounds and the percentage of DPPH (2,2-Diphenyl-1-pikryl-hydrazyl) radical scavenging activity in all the cultivars and selections. Boiling minimized these losses. Red fleshed tubers contained more flavonoids, whereas purple tubers contained more flavonols. Despite severe loss of these compounds due to cooking, both the flesh types retained larger amounts of all these compounds due to higher initial levels. Decline in the radical scavenging activity is directly related to loss of these compounds due to cooking treatments in all white and colored flesh tubers. Red and purple fleshed tubers exhibited greater radical scavenging activity than yellow and white fleshed tubers after each of the cooking treatments. Correction procedures were introduced to exclude interfering compounds (ascorbic acid, other oxidizing agents and reducing sugars) in Folin-Ciocalteu Reagent (FCR) assay of estimating total phenolics in potato.     

Friction and wear behavior of Cu-CeO2 nanocomposite coatings synthesized by pulsed electrodeposition

The friction and wear behavior of copper matrix nanocomposites reinforced with nanosized ceria particles, synthesized by pulse electrocodeposition technique, have been investigated. Tests have been carried out under dry sliding conditions by rubbing against a steel ball at varying loads ranging from 4 to 20 N and at constant speed of 11 rpm using a ball-on-disk wear tester. The experimental results indicate that the wear resistance of copper composite is superior to that of pure copper at all the loads and it improves with the increasing amount of ceria in the copper matrix. The friction coefficient and wear rates increase with the increase in applied load. When the load increases from 4 to 20 N, the transition of wear regime from local damage to delamination of a mechanically mixed layer (MML) occurs.