Numerical Aspects in Locating the Corner of the L-curve
The L-curve method can be used to choose the regularization parameter, when discrete ill-posed problems are solved by a regularization method. In this paper, we propose a novel numerical algorithm for locating the corner of the L-curve when only a finite set of its points is known. The proposed technique is based on conic section fitting. The corner point is chosen to be the nearest one to the shoulder point of the conic. The performance of the method is assessed by comparing its results with those obtained with another approach reported previously by Hansen .
KeywordsSingular Value Decomposition Regularization Method Corner Point Tikhonov Regularization Conic Section
Unable to display preview. Download preview PDF.
- 1.Farin G. (1988), Curves and Surfaces for Computer Aided Geometric Design, Academic Press, New YorkGoogle Scholar
- 2.Guerra V., Hernández V. (1999), Numerical Issues in locating the corner of the L-curve, Technical Report 99–57, Center of Mathematics and Theoretical Physics, HavanaGoogle Scholar
- 3.Grefer H.(1999), An a posteriori parameter choice for ordinary and iterated Tikhonov regularization of ill-posed problems leading to optimal convergence rates. Math. Comput. 49:.507–522Google Scholar
- 7.Hansen P. C. (1998), Regularization Tools. A MATLAB package for Analysis and Solution of Discrete Ill-posed Problems, Technical Report, Dept. of Mathematical Modelling, Technical University of DenmarkGoogle Scholar
- 10.Hernandez V., Estrada J., Barrera P. (1998), On the Euclidean distance from a point to a conic. Revista Integración, Vol 15, No.1: 45–61Google Scholar
- 11.Morozov V. A. (1966), On the solution of functional equations by the method of regularization. Soviet Math. Dokl, 7: 414–41Google Scholar
- 14.Tikhonov A.T., Arsenin V.Y. (1977), Solutions of Ill-Posed Problems, John Wiley, New YorkGoogle Scholar
- 15.Wahba G. (1977), Practical approximate solutions to linear operator equations when the data are noisy. SIAM J. Numer. Anal. 14Google Scholar