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.

Improved Fault Location on Distribution Feeders Based on Matching During-Fault Voltage Sags

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Vortex lattice and matching fields for a long superconducting wire

We investigate the flux penetration patterns and matching fields of a long cylindrical wire of circular cross section in the presence of an external magnetic field. For this study we write the London theory for a long cylinder both for the mixed and Meissner states, with boundary conditions appropriate for this geometry. Using the Monte Carlo simulated annealing method, the free energy of the mixed state is minimized with respect to the vortex position and we obtain the ground state of the vortex lattice for N=3 up to 18 vortices. The free energy of the Meissner and mixed states provides expressions for the matching fields. We find that, as in the case of samples of different geometry, the finite-size effect provokes a delay on the vortex penetration and a vortex accumulation in the center of the sample. The vortex patterns obtained are in good agreement with experimental results.

Three-dimensional reconstruction of fracture surfaces: Area matching algorithms for automatic parallax measurements

This paper describes two solutions for systematic measurement of surface elevation that can be used for both profile and surface reconstructions for quantitative fractography case studies. The first one is developed under Khoros graphical interface environment. It consists of an adaption of the almost classical area matching algorithm, that is based on cross-correlation operations, to the well-known method of parallax measurements from stereo pairs. A normalization function was created to avoid false cross-correlation peaks, driving to the true window best matching solution at each region analyzed on both stereo projections. Some limitations to the use of scanning electron microscopy and the types of surface patterns are also discussed. The second algorithm is based on a spatial correlation function. This solution is implemented under the NIH Image macro programming, combining a good representation for low contrast regions and many improvements on overall user interface and performance. Its advantages and limitations are also presented.

Matching fields of a long superconducting film

We obtain the vortex configurations, the matching fields, and the magnetization of a superconducting film with a finite cross section. The applied magnetic field is normal to this cross section, and we use the London theory to calculate many of its properties, such as the local magnetic field, the free energy, and the induction for the mixed state. Thus previous similar theoretical works, done for an infinitely long superconducting film, are recovered here, in the special limit of a very long cross section. ©1999 The American Physical Society.

Concept of Matching Parallelepiped and its use in the correspondence problem

In this paper, the concept of Matching Parallelepiped (MP) is presented. It is shown that the volume of the MP can be used as an additional measure of `distance' between a pair of candidate points in a matching algorithm by Relaxation Labeling (RL). The volume of the MP is related with the Epipolar Geometry and the use of this measure works as an epipolar constraint in a RL process, decreasing the efforts in the matching algorithm since it is not necessary to explicitly determine the equations of the epipolar lines and to compute the distance of a candidate point to each epipolar line. As at the beginning of the process the Relative Orientation (RO) parameters are unknown, a initial matching based on gradient, intensities and correlation is obtained. Based on this set of labeled points the RO is determined and the epipolar constraint included in the algorithm. The obtained results shown that the proposed approach is suitable to determine feature-point matching with simultaneous estimation of camera orientation parameters even for the cases where the pair of optical axes are not parallel.