Hierarchical classifier design in high-dimensional numerous classcases

In applying pattern recognition methods in remote sensing problems, an inherent limitation is that there is almost always only a small number of training samples with which to design the classifier. A hybrid decision tree classifier design procedure that produces efficient and accurate classifiers for this situation is proposed. In doing so, several key questions are addressed, among them the question of the feature extraction techniques to be used and the mathematical relationship between sample size, dimensionality, and risk value. Empirical tests comparing the hybrid design classifier with a conventional single layered one are presented. They suggest that the hybrid design produces higher accuracy with fewer features. The need for fewer features is an important advantage, because it reflects favorably on both the size of the training set needed and the amount of computation time that will be needed in analysis     

Improved Poisson intensity estimation: denoising application using Poisson data

Recently, Timmermann and Nowak developed algorithms for estimating the means of independent Poisson random variables. The algorithms are based on a multiscale model where certain random variables are assumed to obey a beta-mixture density function. Timmermann and Nowak simplify the density estimation problem by assuming the beta parameters are known and only one mixture parameter is unknown. They use the observed data and the method of moments to estimate the unknown mixture parameter. Taking a different approach, we generate training data from the observed data and compute maximum likelihood estimates of all of the beta-mixture parameters. To assess the improved performance obtained by the proposed modification, we consider a denoising application using Poisson data.     

Traffic grooming on WDM rings using optical burst transport

A sublambda traffic-grooming scheme on wavelength-division-multiplexed (WDM) rings, which is called optical burst transport (OBT), is proposed. The network protocol and architecture allow increased flexibility to tailor the transport network behavior for efficient delivery of bursty data traffic. Using different network parameters, its performance is analyzed via simulation, and the implementation issues including the media-access-control (MAC) protocol, tunable-filter controller, and burst-mode receiver are addressed. Finally, the feasibility of the OBT with an experimental testbed built by the authors is demonstrated and a streaming-video application is used to present its overall functionality.     

Effect of Annealing on the Surface Microstructure of Y-Ba-Cu-O Superconducting Ceramics

Precusors to the YBa₂Cu₃O_7-δ material were prepared by both the oxalate precipitation method and conventional ceramic processing. Second-phase grains were observed to develop on the exposed fracture surface during annealing at 500° to 600°C in an ambient atmosphere. The second-phase grains were identified to be BaCO₃ using EDX and XRD. The liquid phase, which was formed because of local chemical inhomogeneity in the Y-Ba-Cu-O system, contributed to the formation of second-phase grains as a source of Ba.     

Brazing as a means of sealing ceramic membranes for use in advanced coal gasification processes

Coal is potentially a very inexpensive source of clean hydrogen fuel for use in fuel cells, turbines, and various process applications. To realize its potential however, efficient low-cost gas separation systems are needed to provide high purity oxygen that will enhance the coal gasification reaction and to extract hydrogen from the resulting gas product stream. Several types of inorganic membranes are being developed for hydrogen or oxygen separation, including porous alumina, transition metal oxide perovskites, and zirconia. Because they form the heart of the working device, numerous advances have been made in the fabrication and performance of these membrane materials. However, less emphasis has been placed on the materials that will be used in the balance of the device; in particular, the seals that bond the functional ceramic to the metallic structural component. In an effort to begin addressing this issue, we have examined ceramic-to-metal brazing as a method of sealing a model set of gas separation component materials: yttria-stabilized zirconia and stainless steel. In comparative high-temperature exposure testing of joints prepared using commercial brazes and a newly conceived braze alloy, the commercial material proved to be unsuitable due to excessive oxidation. On the other hand, the new material not only displayed superior oxidation resistance, but also excellent hermeticity in prototypic membrane testing.     

Structural properties of 0.65Pb(Mg1/3Nb2/3)O3–0.35PbTiO3 relaxor ferroelectric thin films on SrRuO3 conducting oxides

Coating of 0.65Pb(Mg_1/3Nb_2/3)O₃–0.35PbTiO₃ (PMN–PT) relaxor ferroelectrics by a sol–gel method is followed by growth of epitaxial SrRuO₃ (SRO) metallic oxide electrodes on SrTiO₃ (STO) single-crystal substrate by pulsed laser deposition. High-quality PMN–PT films on SRO with preferred growth orientation were successfully fabricated by controlling the operation parameters. Structural properties of relaxor ferroelectric PMN–PT thin films on SRO/STO substrates have been studied by X-ray diffraction (XRD), transmission electron microscopy (TEM) and atomic force microscopy (AFM). In-plane and out-of-plane alignments of the heterostructure are confirmed and the structural twinning of the materials are also revealed.     

Optical scattering in beef steak to predict tenderness using hyperspectral imaging in the VIS-NIR region

The objective of this research is to develop a non-destructive method for predicting cooked beef tenderness using optical scattering of light on fresh beef muscle tissue. A hyperspectral imaging system (λ = 496–1,036 nm) that consists of a CCD camera and an imaging spectrograph, was used to acquire beef steak images. The hyperspectral image consisted of 120 bands with spectral intervals of 4.54 nm. Sixty-one fresh beef steaks, including 44 strip loin and 17 tenderloin cuts, were collected. After imaging, the steaks were cooked and Warner-Bratzler shear (WBS) force values were collected as tenderness references. The optical scattering profiles were derived from the hyperspectral images and fitted to the modified Lorentzian function. Parameters, such as the peak height, full scattering width at half maximum (FWHM), and the slope around the FWHM were determined at each wavelength. Stepwise regression was used to identify 7 key wavelengths and parameters. The parameters were then used to predict the WBS scores. The model was able to predict WBS scores with an R = 0.67. Optical scattering implemented with hyperspectral imaging shows limited success for predicting current status of tenderness in beef steak.     

Study on Contamination Control for Yield Enhancement in the Manufacturing Line of Cellular Phone Modules

Practical studies on the method of contamination control for yield enhancement in the cellular phone modules production line were carried out. A contamination control method was proposed, consisting of data collection, data analysis, improvement action, verification, and implementation of control. The partition check method and the composition analysis for data collection and data analysis were respectively used in the cellular phone modules manufacturing process, and these methods were evaluated by the variation of yield loss before and after implementing the actions for improvement. In the partition check method, the critical process step was selected, and yield loss reduction through improvement actions was observed, whereas in the composition analysis, critical sources were selected, and yield loss reduction through improvement actions was investigated. From the results, it is concluded that the partition check and the composition analysis are effective solutions for contamination control in cleanroom production lines.