Pattern recognition theory in nonlinear signal processing

A body of research has developed within the context of nonlinear signal and image processing that deals with the automatic, statistical design of digital window-based filters. Based on pairs of ideal and observed signals, a filter is designed in an effort to minimize the error between the ideal and filtered signals. The goodness of an optimal filter depends on the relation between the ideal and observed signals, but the goodness of a designed filter also depends on the amount of sample data from which it is designed. In order to lessen the design cost, a filter is often chosen from a given class of filters, thereby constraining the optimization and increasing the error of the optimal filter. To a great extent, the problem of filter design concerns striking the correct balance between the degree of constraint and the design cost. From a different perspective and in a different context, the problem of constraint versus sample size has been a major focus of study within the theory of pattern recognition. This paper discusses the design problem for nonlinear signal processing, shows how the issue naturally transitions into pattern recognition, and then provides a review of salient related pattern-recognition theory. In particular, it discusses classification rules, constrained classification, the Vapnik-Chervonenkis theory, and implications of that theory for morphological classifiers and neural networks. The paper closes by discussing some design approaches developed for nonlinear signal processing, and how the nature of these naturally lead to a decomposition of the error of a designed filter into a sum of the following components: the Bayes error of the unconstrained optimal filter, the cost of constraint, the cost of reducing complexity by compressing the original signal distribution, the design cost, and the contribution of prior knowledge to a decrease in the error. The main purpose of the paper is to present fundamental principles of pattern recognition theory within the framework of active research in nonlinear signal processing.

Acute swimming stress induces changes in noradrenergic mechanisms in order to maintain the response pattern of the rat vas deferens to norepinephrine

This study investigated mechanisms involved in the maintenance of the functional response pattern of the postjunctional alpha(1)-adrenoceptor in vas deferens isolated from rats submitted to acute swimming stress. The plasma corticosterone levels increased approximately three times after the swimming stress in the nontreated rats as well as after swimming stress in the rats pretreated with desipramine (DMI), yohimbine (YO), or DMI with YO. No alteration was detected in the sensitivity to norepinephrine (NE) in the vasa deferentia from the stressed rats or stressed rats treated with DMI or DMI with YO, in relation to their respective control. However, when the vasa deferentia were previously incubated with DMI, a reduction in sensitivity to NE in organs from stressed rats was observed. Vasa deferentia excised from rats pretreated with YO before the swimming stress showed an increase in postjunctional alpha(1)-response that was abolished by prazosin (PZ). Thus, the neuronal uptake, the prejunctional alpha(2)-adrenoceptors (mediating prejunctional inhibition), the occupancy and functional response of the postjunctional alpha(1)-adrenoceptors, and the emotional stress component were very important for the determination of the noradrenergic response pattern in vas deferens from rats submitted to acute swimming stress. (C) 2002 Elsevier B.V. Ltd. All rights reserved.

Effects of direction and curvature on variable error pattern of reaching movements

A number of studies have analyzed various indices of the final position variability in order to provide insight into different levels of neuromotor processing during reaching movements. Yet the possible effects of movement kinematics on variability have often been neglected. The present study was designed to test the effects of movement direction and curvature on the pattern of movement variable errors. Subjects performed series of reaching movements over the same distance and into the same target. However, due either to changes in starting position or to applied obstacles, the movements were performed in different directions or along the trajectories of different curvatures. The pattern of movement variable errors was assessed by means of the principal component analysis applied on the 2-D scatter of movement final positions. The orientation of these ellipses demonstrated changes associated with changes in both movement direction and curvature. However, neither movement direction nor movement curvature affected movement variable errors assessed by area of the ellipses. Therefore it was concluded that the end-point variability depends partly, but not exclusively, on movement kinematics.

Relationship between carbon source, production and pattern action of alpha-amilase from Rhizopus sp

This paper discusses the inducer effect of corn soluble starch and the individual components (amylose and amylopectin) from corn and potatoes starch for alpha-amylase production by a strain of Rhizopus sp. The following decreasing order in the enzyme production was obtained: corn amylose > potatoes amylose > corn amylopectin > potatoes amylopectin > starch > maltose, coinciding with the ability of the enzyme to release reducing units, except the soluble starch that was more softly hydrolysed. However, when the enzyme action was measured by the iodine binding method, an inverse order of enzyme activity was obtained, that is: amylopectins > starch > amylosis. The results suggest that: a) branched structures in substrate affect the enzyme production; b) corn amylose and corn amylopectin are better inducers than their respectives homologous from potatoes; c) cc-amylase from Rhizopus sp has different action patterns on substrates with straight or branched chains: from the former, it removes only reducing units with lower molecular weight (G1-G3); from the latter it also removes oligosaccharides with higher molecular weight (G5-G6).

Effects of displacement and trajectory length on the variability pattern of reaching movements

The design of the present study enabled the authors to distinguish between the possible effects of movement displacement and trajectory length on the pattern of final positions of planar reaching movements. With their eyes closed, 9 subjects performed series of fast and accurate movements from different initial positions to the same target. For some series, the movements were unconstrained and were therefore performed along an approximately straight vertical line. For other series, an obstacle was positioned so that trajectory length was increased because of an increase in movement curvature. Ellipses of variability obtained by means of principal component analysis applied to the scatter of movement final positions enabled the authors to assess the pattern of movement variable errors. The results showed that the orientation of the ellipses was not affected by movement displacement or by trajectory length, whereas variable errors increased with move ment displacement. An increase in trajectory length as a consequence of increased curvature caused no change in variable error. From the perspective of current motor control theory, that finding was quite unexpected. Further studies are required so that one can distinguish among the possible effects of various kinematics, kinetics, and other variables that could affect the pattern of variable errors of reaching movements.

Oviposition pattern and within-season spatial and temporal variation of pre-dispersal seed predation in a population of Mimosa bimucronata trees

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Bone healing pattern in surgically created circumferential defects around submerged implants: an experimental study in dog

Objective: To describe the healing of marginal defects below or above 1 mm of dimension around submerged implants in a dog model.Material and methods: In 12 Labrador dogs, all mandibular premolars and first molars were extracted bilaterally. After 3 months of healing, full-thickness flaps were elevated in the edentulous region of the right side of the mandible. Two recipient sites were prepared and the marginal 5mm were widened to such an extent to obtain, after implant installation, a marginal gap of 0.5mm at the mesial site (small defect) and of 1.25mm at the distal site (large defect). Titanium healing caps were affixed to the implants and the flaps were sutured allowing a fully submerged healing. The experimental procedures were subsequently performed in the left side of the mandible. The timing of the experiments and sacrifices were planned in such a way to obtain biopsies representing the healing after 5, 10, 20 and 30 days. Ground sections were prepared and histomorphometrically analyzed.Results: The filling of the defect with newly formed bone was incomplete after 1 month of healing in all specimens. Bone formation occurred from the base and the lateral walls of the defects. A larger volume of new bone was formed in the large compared with the small defects. Most of the new bone at the large defect was formed between the 10- and the 20-day period of healing. After 1 month of healing, the outline of the newly formed bone was, however, located at a similar distance from the implant surface (about 0.4mm) at both defect types. Only minor newly formed bone in contact with the implant, starting from the base of the defects, was seen at the large defects (about 0.8mm) while a larger amount was detected at the small defects (about 2.2 mm).Conclusion: Marginal defects around titanium implants appeared to regenerate in 20-30 days by means of a distance osteogenesis. The bone fill of the defects was, however, incomplete after 1 month.

Colonization pattern of Cecropia by Azteca ants: Influence of plant ontogeny, environment and host plant choice by queens

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Aircraft interior failure pattern recognition utilizing text mining and neural networks

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Rabies in southeast Brazil: a change in the epidemiological pattern

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Refractive and geometric lens characterization through multi-wavelength digital speckle pattern interferometry

Physical parameters of different types of lenses were measured through digital speckle pattern interferometry (DSPI) using a multimode diode laser as light source. When such lasers emit two or more longitudinal modes simultaneously the speckle image of an object appears covered of contour fringes. By performing the quantitative fringe evaluation the radii of curvature as well as the refractive indexes of the lenses were determined. The fringe quantitative evaluation was carried out through the four- and the eight-stepping techniques and the branch-cut method was employed for phase unwrapping. With all these parameters the focal length was calculated. This whole-field multi-wavelength method does enable the characterization of spherical and aspherical lenses and of positive and negative ones as well. (C) 2007 Elsevier B.V. All rights reserved.