Examinando por Materia "Information science"
Mostrando 1 - 4 de 4
Resultados por página
Opciones de ordenación
Ítem Kinematic identification of parallel mechanisms by a divide and conquer strategy(INSTICC-INST SYST TECHNOLOGIES INFORMATION CONTROL & COMMUNICATION, 2010-01-01) Durango S.; Externo - Escuela - Ingeniería; Ruiz O.; Restrepo-Giraldo J.; Achiche S.; Universidad EAFIT. Departamento de Ingeniería Mecánica; Laboratorio CAD/CAM/CAEThis paper presents a Divide and Conquer strategy to estimate the kinematic parameters of parallel symmetrical mechanisms. The Divide and Conquer kinematic identification is designed and performed independently for each leg of the mechanism. The estimation of the kinematic parameters is performed using the inverse calibration method. The identification poses are selected optimizing the observability of the kinematic parameters from a Jacobian identification matrix. With respect to traditional identification methods the main advantages of the proposed Divide and Conquer kinematic identification strategy are: (i) reduction of the kinematic identification computational costs, (ii) improvement of the numerical efficiency of the kinematic identification algorithm and, (iii) improvement of the kinematic identification results. The contributions of the paper are: (i) The formalization of the inverse calibration method as the Divide and Conquer strategy for the kinematic identification of parallel symmetrical mechanisms and, (ii) a new kinematic identification protocol based on the Divide and Conquer strategy. As an application of the proposed kinematic identification protocol the identification of a planar 5R symmetrical mechanism is simulated. The performance of the calibrated mechanism is evaluated by updating the kinematic model with the estimated parameters and developing simulations.Ítem A Riemannian geometry in the q-exponential Banach manifold induced by q-divergences(SPRINGER, 2013-01-01) Loaiza, G.; Quiceno, H.R.; Loaiza, G.; Quiceno, H.R.; Universidad EAFIT. Departamento de Ciencias; Matemáticas y AplicacionesFor the family of non-parametric q-exponential statistical models, in a former paper, written by the same authors, a differentiable Banach manifold modelled on Lebesgue spaces of real random variables has been built. In this paper, the geometry induced on this manifold is characterized by q-divergence functionals. This geometry turns out to be a generalization of the geometry given by Fisher information metric and Levi-Civita connections. Moreover, the classical Amari's a-connections appears as special case of the q-connections ?(q). The main result is the expected one, namely the zero curvature of the manifold. © 2013 Springer-Verlag.Ítem Spatial layout and surface reconstruction from omnidirectional images(Institute of Electrical and Electronics Engineers Inc., 2016-01-01) Posada, L.F.; Velasquez-Lopez, A.; Posada, L.F.; Velasquez-Lopez, A.; Universidad EAFIT. Departamento de Ciencias; Ciencias Biológicas y Bioprocesos (CIBIOP)This paper presents a spatial layout recovery approach from single omnidirectional images. Vertical structures in the scene are extracted via classification from heterogeneous features computed at the superpixel level. Vertical surfaces are further classified according to their main orientation by fusing oriented line features, floor-wall boundary features and histogram of oriented gradients (HOG) with a Random Forest classifier. Oriented line features are used to build an orientation map that considers the main vanishing points. The floor-wall boundary feature attempts to reconstruct the scene shape as if it were observed from a bird's-eye view. Finally, the HOG descriptors are aggregated per superpixel and summarize the gradient distribution at homogeneous appearance regions. Compared to existing methods in the literature which rely only on corners or lines, our method gains statistical support from multiple cues aggregated per superpixel which provide more robustness against noise, occlusion, and clutter. © 2016 IEEE.Ítem Spatial layout and surface reconstruction from omnidirectional images(Institute of Electrical and Electronics Engineers Inc., 2016-01-01) Posada, L.F.; Velasquez-Lopez, A.This paper presents a spatial layout recovery approach from single omnidirectional images. Vertical structures in the scene are extracted via classification from heterogeneous features computed at the superpixel level. Vertical surfaces are further classified according to their main orientation by fusing oriented line features, floor-wall boundary features and histogram of oriented gradients (HOG) with a Random Forest classifier. Oriented line features are used to build an orientation map that considers the main vanishing points. The floor-wall boundary feature attempts to reconstruct the scene shape as if it were observed from a bird's-eye view. Finally, the HOG descriptors are aggregated per superpixel and summarize the gradient distribution at homogeneous appearance regions. Compared to existing methods in the literature which rely only on corners or lines, our method gains statistical support from multiple cues aggregated per superpixel which provide more robustness against noise, occlusion, and clutter. © 2016 IEEE.