In this paper, a new method for identifying the dynamical parameters of local constraining supports such as mass, stiffness, and damping was developed through combining the measured frequency transfer functions and st...
详细信息
In this paper, a new method for identifying the dynamical parameters of local constraining supports such as mass, stiffness, and damping was developed through combining the measured frequency transfer functions and structural modification techniques. Since measurement noise often leads to erroneous identifications, regularization techniques have been implemented to reduce noise amplification in the inverse problem. The developed technique has been validated by numerical tests on a multi-supported flexible structure, which can be seen as an idealized electricity generator rotor shaft. The results are satisfactory for noise-free data as well as under realistic noise levels. The sensitivity of the identified support features to noise levels is asserted through a parametric study (C) 2017 Academie des sciences. Published by Elsevier Masson SAS. All rights reserved.
We consider the problem of image segmentation in magnetic resonance imaging (MRI) directly from k-space measurements as opposed to more standard approaches that first reconstruct the image before segmenting it. The mo...
详细信息
ISBN:
(纸本)9783030223687;9783030223670
We consider the problem of image segmentation in magnetic resonance imaging (MRI) directly from k-space measurements as opposed to more standard approaches that first reconstruct the image before segmenting it. The model we employ is the piecewise-constant Mumford-Shah model (the Potts model) in connection with the MRI reconstruction problem. The output of our proposed scheme is a piecewise-constant function which yields a segmentation of the image domain. To solve the involved non-convex and non-smooth optimization problem, we adopt an iterative minimization strategy based on surrogate functionals. Numerical experiments illustrate the potential of the approach when applied on undersampled MRI data.
暂无评论