Scene segmentation using neuromorphic oscillatory networks

Using the neuromorphic approach, we propose an analog very large-scale integration (VLSI) implementation of an oscillatory segmentation algorithm based on local excitatory couplings and global inhibition. The original model has been simplified and adapted for its efficient VLSI implementation while preserving its segmentation properties. To demonstrate the feasibility of the approach, a 16/spl times/16-pixel testchip has been manufactured. Extensive experimental results demonstrate that it can properly segment binary images. Power consumption, segmentation time per cell, and system complexity are very low compared to other hardware and software implementation schemes. We also show two main differences between the original algorithm and the analog approach. First, the network is noise tolerant without the need of additional elements and second, delays between oscillators due to the combination of mismatch and output capacitances have to be accounted for network performance.     

Diversity training in graduate school: An exploratory evaluation of the Safe Zone project.

How can psychologists and graduate students become more affirmative of lesbian, gay, bisexual, and transgender (LGBT) issues? Safe Zone is a diversity-training program to increase sensitivity toward, knowledge of, and advocacy for LGBT populations and issues that affect them. In this exploratory study, the implementation of Safe Zone is described within a school of professional psychology; its effectiveness was assessed by self-reported behavioral and attitudinal changes and from participant evaluations. Results were encouraging with regard to the goals of Safe Zone and its adaptation to institutions interested in improving relations with the LGBT community, but additional empirical studies are needed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Results of benchmark calculations based on OLHF-1 test

The subject of this paper is to compare the results of the different calculations performed by the benchmark participants in the framework of the OECD Lower Head Failure (OLHF) program. The benchmark consists in the finite element (FE) calculation or in analytical calculations of the mechanical behavior of the OLHF-1 experiment. Seven participants from six countries and seven companies or organizations (AVN, VTT, GRS, UJV, SNL, IPSN and CEA) have performed the benchmark.The OLHF experiment program extends the NRC-sponsored SNL LHF program (NUREG/CR-5582) completed in 1998: these experiments where intended to simulate the thermal/mechanical loads to a 1/4.85-scale model of a reactor pressure vessel. The pressure vessel material (SA533B1 steel) used in these experiments is prototypic of reactor PWR vessel material and has been well characterized by material property testing as part of this program. The OLHF tests advance the results of the previous testing program by examining the effects of large temperature differences across the vessel wall. Large temperature differences in excess of 150–400 K are more prototypic of accident conditions.Most of the participants performed 2-D axisymmetric analyses and doesn’t study the crack opening. The global mechanical behaviour of OLHF-1 experiment is well represented but prediction of the maximum vertical displacement is not in good agreement with the experimental value. Failure time and location are in quite good agreement with experimental results but large discrepancies are observed on the mode of failure: creep or plasticity. To improve predictions, more investigation and work is needed on the choice of the failure criteria and failure mode.     

A VLSI architecture for a half-edge-based corner detector

Corner detection is a low-level feature detection operator that is of great use in image processing applications, for example, optical flow and structure from motion by image correspondence. The detection of corners is a computationally intensive operation. Past implementations of corner detection techniques have been restricted to software. In this paper we propose an efficient very large-scale integration (VLSI) architecture for detection of corners in images. The corner detection technique is based on the half-edge concept and the first directional derivative of Gaussian. Apart from the location of the corner points, the algorithm also computes the corner orientation and the corner angle and outputs the edge map of the image. The symmetrical properties of the masks are utilized to reduce the number of convolutions effectively, from eight to two. Therefore, the number of multiplications required per pixel is reduced from 1800 to 392. Thus, the proposed architecture yields a speed-up factor of 4.6 over conventional convolution architectures. The architecture uses the principles of pipelining and parallelism and can be implemented in VLSI.     

Single-Phase Permanent-Magnet Motor Analysis

A unified theoretical approach is presented for the calculation of the performance of the single-phase capacitor-start permanent-magnet (PM) ac line-start motor. The earlier work of Chang and of Finch and Lawrenson on capacitor reluctance motors is extended. As far as possible, a common set of reference frame transformations for steady-state, transient, and asynchronous performance is also provided, permitting the use of all the theory developed for the analysis of the balanced polyphase PM motor. A new approximate technique is given for estimating the average asynchronous torque/speed curve from computed acceleration curves.     

Investigation of the impact of changes in demand forecasting method on the financial performance of an electricity supply undertaking

Peak demand forecasts obtained from six different univariate forecasting methods, under a range of conditions, were used to drive a capacity acquisition model of a large electrical supply system; and the resulting physical and financial performance of the model was observed for each set of forecasts. The results obtained are discussed in the context of their implications for the choice of load forecasting method used in capacity acquisition planning by a power supply undertaking.     

High-gain pseudomorphic InGaAs base ballistic hot-electron device

A high-gain ballistic hot-electron device is described. The GaAs-AlGaAs heterostructure device, with a 21-mm-thick pseudomorphic In _0.12Ga_0.88As base, had a current gain of 27 at 77 K and 41 at 4.2 K. As characteristically seen in ballistic devices, transfer into the L valley limited the maximum gain. The Γ-L valley separation in the strained In_0.12Ga_0.88As was estimated to be about 380 meV     

Physical modeling of gas/liquid mass transfer in a gas stirred ladle

The absorption of gas through the plume eye and of an injected gas in a steelmaking ladle process was investigated, using a physical model of CO₂ absorption into a NaOH solution. The results show that the inert gas escaping through the plume eye is ineffective in protecting the bath from the atmosphere, and placing an oil layer (simulated slag) decreases the absorption rate significantly. Increasing the flow rate of the inert gas not only exposes more of the liquid surface to the CO₂ atmosphere, but also increases the mass transfer coefficient at the surface. The overall mass transfer between an injected CO₂ gas and NaOH solution includes the mass transfer through the surface of the bath as well as the mass transfer in the bubble dispersion zone. The difference between the mass transfer in the bubble dispersion zone and the overall mass transfer was found to be significant for relatively low gas flow rates. The mass transfer coefficient of CO₂ in the bubble dispersion zone was estimated using available information regarding the bubble size and velocity. Mass transfer coefficient estimated for the constant bubble frequency regime shows a dependence on gas flow rate. However, if a constant characteristic size of bubbles is assumed as an alternative approach, the mass transfer coefficient is independent of the gas flow rate.     

Pulse stabilisation in harmonically FM modelocked erbium fibre ring laser with DFBLD

A new pulse stabilisation scheme for a harmonically frequency modulated (FM) modelocked erbium fibre ring laser using a distributed feedback laser diode (DFBLD) is presented. More than 65 dB supermode noise suppression in the RF spectrum was obtained with the proposed method. Effective reduction of supermode noise can be accomplished as the DFBLD in the fibre laser acts not only as a fast-gain saturated medium but also as an optical bandpass filter for suppressing supermode noise in a harmonically modelocked laser system.     

Design and fabrication of RESURF MOSFETs on 4H-SiC(0001), (112~0), and 6H-SiC(0001)

Design and fabrication of lateral SiC reduced surface field (RESURF) MOSFETs have been investigated. The doping concentration (dose) of the RESURF and lightly doped drain regions has been optimized to reduce the electric field crowding at the drain edge or in the gate oxide by using device simulation. The optimum oxidation condition depends on the polytype: N/sub 2/O oxidation at 1300/spl deg/C seems to be suitable for 4H-SiC, and dry O/sub 2/ oxidation at 1250/spl deg/C for 6H-SiC. The average inversion-channel mobility is 22, 78, and 68 cm/sup 2//Vs for 4H-SiC(0001), (112~0), and 6H-SiC(0001) MOSFETs, respectively. RESURF MOSFETs have been fabricated on 10-/spl mu/m-thick p-type 4H-SiC(0001), (112~0), and 6H-SiC(0001) epilayers with an acceptor concentration of 1/spl times/10/sup 16/ cm/sup -3/. A 6H-SiC(0001) RESURF MOSFET with a 3-/spl mu/m channel length exhibits a high breakdown voltage of 1620 V and an on-resistance of 234 m/spl Omega//spl middot/cm/sup 2/. A 4H-SiC(112~0) RESURF MOSFET shows the characteristics of 1230 V-138 m/spl Omega//spl middot/cm/sup 2/.