Free Online Productivity Tools
i2Speak
i2Symbol
i2OCR
iTex2Img
iWeb2Print
iWeb2Shot
i2Type
iPdf2Split
iPdf2Merge
i2Bopomofo
i2Arabic
i2Style
i2Image
i2PDF
iLatex2Rtf
Sci2ools

DICTA

2003

2003

An algebraic curve is deﬁned as the zero set of a multivariate polynomial. We consider the problem of ﬁtting an algebraic curve to a set of vectors given an additional set of vectors labelled as interior or exterior to the curve. The problem of ﬁtting a linear curve in this way is shown to lend itself to a support vector representation, allowing non-linear curves and high dimensional surfaces to be estimated using kernel functions. The approach is attractive due to the stability of solutions obtained, the range of functional forms made possible (including polynomials), and the potential for applying well understood regularisation operators from the theory of Support Vector Machines. 1 Motivation Algebraic curves provide a powerful basis for a range of geometrical analysis problems, including shape recognition and non-iterative shape registration, largely due to the capacity for deriving geometric invariants [5, 3, 6]. Geometric invariants are those properties of an algebraic curv...

Related Content

Added |
31 Oct 2010 |

Updated |
31 Oct 2010 |

Type |
Conference |

Year |
2003 |

Where |
DICTA |

Authors |
Christian J. Walder, Brian C. Lovell, Peter J. Kootsookos |

Comments (0)