A Mixed Mode Neural Network Circuitry for Object Recognition Application

A general purpose Conic Section Function Neural Network (CSFNN) circuitry in Very Large Scale Integration (VLSI) has been designed for an object recognition application. CSFNN is capable of making open and closed decision regions by combining the propagation rules of Radial Basis Functions (RBF) and Multilayer Perceptrons (MLP) on a single neural network with a unique propagation rule. Chip-in-the-loop learning technique was used during the training process. Utilizing mixed-mode hardware techniques, the inputs of the network and the feedforward signals are all analog while the control unit and storage of the network parameters are fully digital. CSFNN circuitry architecture is problem independent and consists of 16 inputs, 16 hidden layer neurons and 8 outputs. Inheriting the merits of CSFNN, the circuitry has good recognition performance on several objects with invariance to pose, lighting, and brightness. The designed hardware achieved a good recognition performance by means of both accuracy and computational time comparable to CSFNN software.     

Magnetic resonance image analysis by information theoretic criteriaand stochastic site models

Quantitative analysis of magnetic resonance (MR) images is a powerful tool for image-guided diagnosis, monitoring, and intervention. The major tasks involve tissue quantification and image segmentation where both the pixel and context images are considered. To extract clinically useful information from images that might be lacking in prior knowledge, the authors introduce an unsupervised tissue characterization algorithm that is both statistically principled and patient specific. The method uses adaptive standard finite normal mixture and inhomogeneous Markov random field models, whose parameters are estimated using expectation-maximization and relaxation labeling algorithms under information theoretic criteria. The authors demonstrate the successful applications of the approach with synthetic data sets and then with real MR brain images     

Determination of Boza's Organic Acid Composition as it is Affected by Raw Material and Fermentation

In the current study boza, a low-alcohol-content beverage, is produced from rice, maize, millet and wheat flours by spontaneous fermentation at 24 ± 1°C for 24 h. During the fermentation and storage period, changes in total titritable acidity, pH, alcohol and organic acid profiles, and sensory properties were determined. The major organic acids were found as oxalic, lactic, pyruvic, and acetic acids. Even though the ratings of sensory parameters were not within wide limits, they were significantly affected by storage time. Maize boza scored higher due to the similarity in its colour and flavour to that sold in the market.     

Building community information systems: the Connected Kids case

New technologies make it feasible and, in many cases, practical for individuals, groups, and organizations to collaborate in the development of joint information systems. Users find collaborative information systems attractive because they make it possible to find information from diverse sources easily and efficiently. Such systems also make good sense for information providers because they can attract and serve a larger audience than a solitary effort might otherwise command, making it possible to pool resources to achieve certain economies of scale and technology expense. Conceived as both a digital government project and a community information system, Connected Kids helped bridge one city's digital divide by providing new information technologies to youth organizations.     

Selecting the optimal shift numbers using fuzzy control model: a paint factory’s facility application

Number of shifts is one of the most important criteria for the production planners to minimize the production costs. The first measurement of the facility, where we executed the application, is optimizing the manpower rotation in B12 production, a special kind of chemical, which is used for paint binders. Seasonal variations in demand of B12 add more importance to the number of shifts in the facility. The purpose of this article is to optimize the shift periods subject to raw material, shipping date, inventory and demand constraints. The model is developed for a paint factory’s reactor facility. The production system has a raw material inventory, a reactor, and a finished goods inventory. We use fuzzy control to optimize the number of shifts under the constraints, given before.     

Guest Editor's Introduction

Journal of Signal Processing Systems -     

Optimal design of hybrid laminates with discrete ply angles for maximum buckling load and minimum cost

The design of hybrid symmetric laminated plates consisting of high-stiffness surface and low-stiffness core layers is presented. In the first problem the maximization of buckling load is carried out over a discrete set of ply angles. In the second problem the minimum number of high-stiffness plies is determined for a given buckling load to minimize the material cost. Boolean variables are introduced to specify stacking sequence. Solution of the linear optimization problem yields an optimal stacking sequence. The effect of hybridization is investigated for various problem parameters such as the aspect ratio of the laminate and the number of plies. The optimal designs are obtained with upper bound constraints on the effect of bending-twisting coupling stiffnesses. Results are given for hybrid graphite-epoxy/glass-epoxy laminates under both uniaxial and biaxial loadings.     

Complex ICA by negentropy maximization.

In this paper, we use complex analytic functions to achieve independent component analysis (ICA) by maximization of non-Gaussianity and introduce the complex maximization of non-Gaussianity (CMN) algorithm. We derive both a gradient-descent and a quasi-Newton algorithm that use the full second-order statistics providing superior performance with circular and noncircular sources as compared to existing methods. We show the connection among ICA methods through maximization of non-Gaussianity, mutual information, and maximum likelihood (ML) for the complex case, and emphasize the importance of density matching for all three cases. Local stability conditions are derived for the CMN cost function that explicitly show the effects of noncircularity on convergence and demonstrated through simulation examples.     

A filtering approach for reducing timing jitter due to the acousticeffect

We introduce a signal processing approach to compensate for the timing jitter produced by the acoustic effect in soliton communications. The other main sources of timing jitter, the Gordon-Haus effect and the polarization effect, are inherently stochastic. By contrast, the acoustic effect is deterministic and becomes the dominant source of bit error rates in standard soliton systems when the bit rates are more than 10 Gb/s and the transmission distance is more than several thousand kilometers. We exploit the deterministic nature of the acoustic effect to introduce a scheme that predicts the amount of timing jitter as a function of the previous transmitted bits and uses the information to adjust the sampling period of the received soliton pulses. We demonstrate successful application of the scheme by simulations and discuss implementation issues     

Quantitative Analysis of MR Brain Image Sequences by Adaptive Self-Organizing Finite Mixtures

This paper presents an adaptive structure self-organizing finite mixture network for quantification of magnetic resonance (MR) brain image sequences. We present justification for the use of standard finite normal mixture model for MR images and formulate image quantification as a distribution learning problem. The finite mixture network parameters are updated such that the relative entropy between the true and estimated distributions is minimized. The new learning scheme achieves flexible classifier boundaries by forming winner-takes-in probability splits of the data allowing the data to contribute simultaneously to multiple regions. Hence, the result is unbiased and satisfies the asymptotic optimality properties of maximum likelihood. To achieve a fully automatic quantification procedure that can adapt to different slices in the MR image sequence, we utilize an information theoretic criterion that we have introduced recently, the minimum conditional bias/variance (MCBV) criterion. MCBV allows us to determine the suitable number of mixture components to represent the characteristics of each image in the sequence. We present examples to show that the new method yields very efficient and accurate performance compared to expectation-maximization, K-means, and competitive learning procedures.