Characterization of texture in image of skin lesions by support vector machine

Due to the increased incidence of skin cancer, computational methods based on intelligent approaches have been developed to aid dermatologists in the diagnosis of skin lesions. This paper proposes a method to classify texture in images, since it is an important feature for the successfully identification of skin lesions. For this is defined a feature vector, with the fractal dimension of images through the box-counting method (BCM), which is used with a SVM to classify the texture of the lesions in to non-irregular or irregular. With the proposed solution, we could obtain an accuracy of 72.84%. © 2012 AISTI.

Measuring and analyzing color and texture information in anatomical leaf cross sections: An approach using computer vision to aid plant species identification

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

Reconhecimento de contorno de edifício em imagens de alta resolução usando os momentos complexos de Zernike

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

Extração de feições retas com precisão subpixel: estudo comparativo entre três métodos

In the fields of Machine Vision and Photogrammetry, extracted straight lines from digital images can be used either as vector elements of a digital representation or as control entities that allow the determination of the camera interior and exterior orientation parameters. Applications related with image orientation require feature extraction with subpixel precision, to guarantee the reliability of the estimated parameters. This paper presents three approaches for straight line extraction with subpixel precision. The first approach considers the subpixel refinement based on the weighted average of subpixel positions calculated on the direction perpendicular to the segmented straight line. In the second approach, a parabolic function is adjusted to the grey level profile of neighboring pixels in a perpendicular direction to the segmented line, followed by an interpolation of this model to estimate subpixel coordinates of the line center. In the third approach, the subpixel refinement is performed with a parabolic surface adjustment to the grey level values of neighboring pixels around the segmented line. The intersection of this surface with a normal plane to the line direction generates a parabolic equation that allows estimating the subpixel coordinates of the point in the straight line, assuming that this is the critical point of this function. Three experiments with real images were made and the approach based on parabolic surface adjustment has presented better results.

Resonances and subharmonic bifurcations of large amplitude periodic orbits of planar polynomial vector fields

In this work are studied periodic perturbations, depending on two parameters, of planar polynomial vector fields having an annulus of large amplitude periodic orbits, which accumulate on a symmetric infinite heteroclinic cycle. Such periodic orbits and the heteroclinic trajectory can be seen only by the global consideration of the polynomial vector fields on the whole plane, and not by their restriction to any compact set. The global study involving infinity is performed via the Poincare Compactification. It is shown that, for certain types of periodic perturbations, one can seek, in a neighborhood of the origin in the parameter plane, curves C-(m) of subharmonic bifurcations, for which the periodically perturbed system has subharmonics of order m, for any integer m.

Periodic perturbations of quadratic planar polynomial vector fields

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Attraction of the cutaneous leishmaniasis vector Nyssomyia neivai (Diptera: Psychodidae) to host odour components in a wind tunnel

Background: Laboratory studies of host-seeking olfactory behaviour in sandflies have largely been restricted to the American visceral leishmaniasis vector Lutzomyia longipalpis. In comparison, almost nothing is known about the chemical ecology of related species, which transmit American cutaneous leishmaniasis (ACL), due in part to difficulties in raising these insects in the laboratory. Understanding how ACL vectors locate their hosts will be essential to developing new vector control strategies to combat this debilitating disease.Methods: This study examined host-odour seeking behaviour of the ACL vector Nyssomyia neivai (Pinto) (=Lutzomyia neivai) using a wind tunnel olfactometer. The primary aim was to determine whether field-collected female N. neivai would respond to host odours in the laboratory, thereby eliminating the need to maintain colonies of these insects for behavioural experiments. Responses to two key host odour components, 1-octen-3-ol and lactic acid, and a commercially-available mosquito lure (BG-Lure (TM)) were assessed and compared relative to an air control. We also tested whether trials could be conducted outside of the normal evening activity period of N. neivai without impacting on fly behaviour, and whether the same flies could be used to assess baseline responses to air without affecting responses to octenol, thereby reducing the number of flies required for experiments.Results: Octenol was found to both activate host-seeking behaviour and attract female N. neivai in the wind tunnel, while lactic acid elicited weaker responses of activation and attractiveness under identical conditions. The BG-Lure did not activate or attract N. neivai under test conditions. Further experiments showed that sandfly behaviour in the wind tunnel was not affected by time of day, such that experiments need not be restricted to nocturnal hours. Moreover, using the same flies to measure both baseline responses to air and attraction to test compounds did not affect odour-seeking behaviour.Conclusions: The results of this study demonstrate that N. neivai taken from the field are suitable for use in laboratory olfactometer experiments. It is hoped this work will facilitate further research into chemical ecology of this species, and other ACL vectors.

A novel algorithm for feature selection using Harmony Search and its application for non-technical losses detection

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

An Optimum-Path Forest framework for intrusion detection in computer networks

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

HOW FAR do WE GET USING MACHINE LEARNING BLACK-BOXES?

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

CENSUS HISTOGRAMS: A SIMPLE FEATURE EXTRACTION and MATCHING APPROACH FOR FACE RECOGNITION

Most face recognition approaches require a prior training where a given distribution of faces is assumed to further predict the identity of test faces. Such an approach may experience difficulty in identifying faces belonging to distributions different from the one provided during the training. A face recognition technique that performs well regardless of training is, therefore, interesting to consider as a basis of more sophisticated methods. In this work, the Census Transform is applied to describe the faces. Based on a scanning window which extracts local histograms of Census Features, we present a method that directly matches face samples. With this simple technique, 97.2% of the faces in the FERET fa/fb test were correctly recognized. Despite being an easy test set, we have found no other approaches in literature regarding straight comparisons of faces with such a performance. Also, a window for further improvement is presented. Among other techniques, we demonstrate how the use of SVMs over the Census Histogram representation can increase the recognition performance.

New Insights on Nontechnical Losses Characterization Through Evolutionary-Based Feature Selection

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

On 3-Parameter Families of Piecewise Smooth Vector Fields in the Plane

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

Bounded solutions of fermions in the background of mixed vector-scalar inversely linear potentials

The problem of a fermion subject to a general mixing of vector and scalar potentials in a two-dimensional world is mapped into a Sturm-Liouville problem. Isolated bounded solutions are also searched. For the specific case of an inversely linear potential, which gives rise to an effective Kratzer potential in the Sturm-Liouville problem, exact bounded solutions are found in closed form. The case of a pure scalar potential with their isolated zero-energy solutions, already analyzed in a previous work, is obtained as a particular case. The behavior of the upper and lower components of the Dirac spinor is discussed in detail and some unusual results are revealed. The nonrelativistic limit of our results adds a new support to the conclusion that even-parity solutions to the nonrelativistic one-dimensional hydrogen atom do not exist. (c) 2004 Elsevier B.V. All rights reserved.

Effects of a mixed vector-scalar screened coulomb potential for spinless particles

The problem of a spinless particle subject to a general mixing of vector and scalar screened Coulomb potentials in a two-dimensional world is analyzed and its bounded solutions are found. Some unusual results, including the existence of a bona fide solitary zero-eigenmode solution, are revealed for the Klein-Gordon equation. The cases of pure vector and scalar potentials, already analyzed in previous works, are obtained as particular cases.