A nonparametric statistical method for image segmentation using information theory and curve evolution.

In this paper, we present a new information-theoretic approach to image segmentation. We cast the segmentation problem as the maximization of the mutual information between the region labels and the image pixel intensities, subject to a constraint on the total length of the region boundaries. We assume that the probability densities associated with the image pixel intensities within each region are completely unknown a priori, and we formulate the problem based on nonparametric density estimates. Due to the nonparametric structure, our method does not require the image regions to have a particular type of probability distribution and does not require the extraction and use of a particular statistic. We solve the information-theoretic optimization problem by deriving the associated gradient flows and applying curve evolution techniques. We use level-set methods to implement the resulting evolution. The experimental results based on both synthetic and real images demonstrate that the proposed technique can solve a variety of challenging image segmentation problems. Futhermore, our method, which does not require any training, performs as good as methods based on training.     

Investigation into the interference potential of spread-spectrum clock generation to broadband digital communications

Since its inception, spread-spectrum clocks have been a valuable technology for the purposes of EMI reduction and EMC compliance. This study evaluates the interference potential of spread spectrum clocks to broadband digital communications such as high definition television (HDTV). This is accomplished by comparing the spread spectrum clock interference susceptibility of digital television (DTV) receivers to the interference susceptibility of analog television receivers. This study shows that DTV receivers are more immune to the same level of interference than existing analog television receivers by 16 dB. Since industry has shown that analog television has had negligible interference from information technology equipment, including spread spectrum clocks, then the digital systems should have even less.     

Phase-shifted Bragg grating filters in polymer waveguides

We fabricated a very narrow-band 0.03-nm (5-GHz) phase-shifted Bragg grating filter at 1290 nm on an LD-3 polymer ridge waveguide. The grating was post processed onto the waveguide by photobleaching using a 457-nm argon ion laser with a phase mask. We present the details of the fabrication process and a measurement of the filter spectrum with an analysis.     

Progress towards using the speed of ultrasound for beef leanness classification

Following the observation of a high correlation between the reciprocal of the speed of ultrasound and lean proportion in beef carcasses, commercial equipment (VOSI) was developed and its performance alongside a smaller assembly (Mini G) is reported. The speed of ultrasound was measured through six sites on beef carcasses within 1 h of stunning: three sites with VOSI and three sites with Mini G. Sides were also scored for fatness and conformation using the EAAP system, and subcutaneous fat depth was measured at 25,50 and 75% of the width of M. longissimus thoracis from the midline at the last rib. All sides were dissected into subcutaneous fat, intermuscular fat, muscle and bone. The reciprocal speed averaged over the three VOSI sites (RV346) was the best single predictor of the proportion of lean in the side. Multiple regression analyses, in predicting per cent lean, showed that addition of conformation score improved the precision of fat score and fat depth for a group of 26 sides. It did not provide a significant improvement over RV346. However, in another group of 50 sides showing more variation in side mass and conformation, conformation score did significantly improve RV346. The best prediction of per cent lean, involving RV346 and tissue thicknesses, gave a residual standard deviation of 1·35%, and conformation score did not significantly improve this regression.     

Thin film Y-Ba-Cu-O/Ag composites for fluxonic devices

We have investigated thin film composites of YBa₂Cu₃O₇ (YBCO) with Ag for fluxonic device applications. YBCO/Ag composite films are produced by first depositing a layer of Ag onto a substrate and then heating the film to the YBCO deposition temperature of 680°C or higher. YBCO is deposited by off-axis sputtering onto the Ag-coated substrate. The resulting YBCO/Ag film is a composite of YBCO with well-defined Ag regions several microns in size. Scanning electron micrograph images of the films' surfaces show a background of smooth YBCO grains dotted with Ag clusters. For a wide range of increasing Ag composition, the transition temperatures of the composite films on SrTiO₃ remain high, while the critical current densities have been reduced as much as 65 times. On MgO substrates, critical current density has been reduced by more than four orders of magnitude. Also on MgO, significant voltage response is seen in external magnetic fields of less than 1 mT. These measurements suggest that the films may be arrays of superconductor-normal-superconductor (SNS) junctions formed by weakly coupled YBCO grains with Ag in the grain boundaries. The field responsivity and low critical current densities of these composites make them potentially useful for fabrication of fluxonic devices     

Learning from Examples with Information Theoretic Criteria

This paper discusses a framework for learning based on information theoretic criteria. A novel algorithm based on Renyi's quadratic entropy is used to train, directly from a data set, linear or nonlinear mappers for entropy maximization or minimization. We provide an intriguing analogy between the computation and an information potential measuring the interactions among the data samples. We also propose two approximations to the Kulback-Leibler divergence based on quadratic distances (Cauchy-Schwartz inequality and Euclidean distance). These distances can still be computed using the information potential. We test the newly proposed distances in blind source separation (unsupervised learning) and in feature extraction for classification (supervised learning). In blind source separation our algorithm is capable of separating instantaneously mixed sources, and for classification the performance of our classifier is comparable to the support vector machines (SVMs).     

Amorphous Boron Nitride: A Universal,Ultrathin Dielectric For 2D Nanoelectronics

Next‐generation nanoelectronics based on 2D materials ideally will require reliable, flexible, transparent, and versatile dielectrics for transistor gate barriers, environmental passivation layers, capacitor spacers, and other device elements. Ultrathin amorphous boron nitride of thicknesses from 2 to 17 nm, described in this work, may offer these attributes, as the material is demonstrated to be universal in structure and stoichiometric chemistry on numerous substrates including flexible polydimethylsiloxane, amorphous silicon dioxide, crystalline Al₂O₃, other 2D materials including graphene, 2D MoS₂, and conducting metals and metal foils. The versatile, large area pulsed laser deposition growth technique is performed at temperatures less than 200 °C and without modifying processing conditions, allowing for seamless integration into 2D device architectures. A device‐scale dielectric constant of 5.9 ± 0.65 at 1 kHz, breakdown voltage of 9.8 ± 1.0 MV cm^?1, and bandgap of 4.5 eV were measured for various thicknesses of the ultrathin a‐BN material, representing values higher than previously reported chemical vapor deposited h‐BN and nearing single crystal h‐BN.     

The wall-jet electrode and the study of electrode reaction mechanisms: the EC reaction

The theory of EC reactions at a wall-jet electrode is developed using a computational procedure based on the Backwards Implicit Method. In particular, for the case of a reversible electron transfer, it is shown that the variation of the halfwave potential with solution flow rate provides a means of characterizing the EC mechanism. A working curve is presented which permits the analysis of experimental data and the deduction of the rate constant for the following chemical reaction.Nomenclature a diameter of the jet - a _j matrix element (j = 2,3,..., ) - b _j matrix element (j = 1,2,..., ) - C _j matrix element (j = 1,2,..., ) - D diffusion coefficient - d _j vector element (j = 1,2,..., ) - E electrode potential - E ⁰ standard electrode potential of A/B couple - F Faraday constant - g(A) normalized concentration of A - g(B) normalized concentration of B - I current - I _LIM transport limited curren - J number of points inr-direction on finite difference grid - j conter,j=1, 2,...,J - K Number of points inz-direction on finite difference grid - K _j,k+1 normalized rate constant (Equation 23) - k counter,k=1, 2,...,K - k counter,k=1, 2,...,K - k _c experimental constant - first order rate constant - M constant defined byM =k _c ⁴ V _f ³ /2³a² - R electrode radius - r radial coordinate - u _j vector element (j = 1,2,..., ) - V volume flow rate (cm³ s^–1) - _r solution velocity inr-direction - _z solution velocity inz-direction - z coordinate normal to electrode surface - dimensionless distance parameter (see Equation 5) - dimensionless variable (Equation 25) - normalized electrode potential - _1/2 normalized halfwave potential - kinematic viscosity (cm² s^–1) - dimensionless variable (Equation 26) - {d} _k vector{d _1,k ,d _2,k ,...d _j–1,k} - {u} _k vector {u _1,k+1,u _2,k+1,...,u _j–1,k+1}