The authors propose an advanced synthetic aperture radar (SAR) image formation framework based on iterative inversion algorithms that approximately solve a regularised least squares problem. The framework provides imp...
详细信息
The authors propose an advanced synthetic aperture radar (SAR) image formation framework based on iterative inversion algorithms that approximately solve a regularised least squares problem. The framework provides improved image reconstructions, compared to the standard methods, in certain imaging scenarios, for example when the SAR data are under-sampled. iterativealgorithms also allow prior information to be used to solve additional problems such as the correction of unknown phase errors in the SAR data. However, for an iterativeinversion framework to be feasible, fast algorithms for the generative model and its adjoint must be available. The authors demonstrate how fast, N-2 log(2) N complexity, (re/back)-projection algorithms can be used as accurate approximations for the generative model and its adjoint, without the limiting geometric approximations of other N-2 log(2) N methods, for example, the polar format algorithm. Experimental results demonstrate the effectiveness of their framework using publicly available SAR datasets.
Electrical Impedance Tomography (EIT) has shown great potential to be used as a method for breast cancer screening. Combining EIT with x-ray mammography allows for both sets of data to be used in conjunction to make a...
详细信息
ISBN:
(纸本)9781612848266
Electrical Impedance Tomography (EIT) has shown great potential to be used as a method for breast cancer screening. Combining EIT with x-ray mammography allows for both sets of data to be used in conjunction to make a more accurate diagnosis. Previous efforts to implement EIT in the mammography geometry have used analytical solutions to the forward model. This paper presents a three dimensional numerical forward model which uses a finite element method (FEM) to implement the complete electrode model (CEM) in EIT. We show that the numerical solution achieves greater accuracy than analytic models, and opens the way to implement an iterative inversion algorithm.
An inversioniterativealgorithm for discrete-time non-linear systems is presented. The algorithm is tuned depending on system characteristics. Under the assumption that the system inverse exists, the iterative algori...
详细信息
An inversioniterativealgorithm for discrete-time non-linear systems is presented. The algorithm is tuned depending on system characteristics. Under the assumption that the system inverse exists, the iterativealgorithm constructs the system inverse by attempting to achieve on-line the perfect tracking. The inversion process is based on the observation of the input-output pairs. The resulting control scheme requires a very low a priori knowledge of the system dynamics. Moreover, low-engineering efforts are needed to apply the control scheme to a particular control problem or to modify it in order to accommodate changes in the physical system. Copyright (C) 2001 John Wiley & Sons, Ltd.
The problem of inverting trained feedforward neural networks is to find the inputs which yield a given output, In general, this problem is an ill-posed problem because the mapping from the output space to the input sp...
详细信息
The problem of inverting trained feedforward neural networks is to find the inputs which yield a given output, In general, this problem is an ill-posed problem because the mapping from the output space to the input space is a one-to-many mapping. In this paper, we present a method for dealing with the inverse problem by using mathematical programming techniques. The principal idea behind the method is to formulate the inverse problem as a nonlinear programming (NLP) problem, a separable programming (SP) problem, or a linear programming (LP) problem according to the architectures of networks to be inverted or the types of network inversions to be computed. An important advantage of the method over the existing iterative inversion algorithm is that various designated network inversions of multilayer perceptrons (MLP's) and radial basis function (RBF) neural networks can be obtained by solving the corresponding SP problems, which can be solved by a modified simplex method, a well-developed and efficient method for solving LP problems. We present several examples to demonstrate the proposed method and the applications of network inversions to examining and improving the generalization performance of trained networks. The results show the effectiveness of the proposed method.
暂无评论