Paper Info
Title
A precise ellipse fitting method for noisy data
Abstract
Least squares based ellipse detection is used as a core process in many image processing applications. In order to restrict the solution to ellipses and avoid non-elliptic conics, constrained optimization has to be incorporated in the least squares model. This paper proposes a least squares method that does not require a constrained optimization and has very low false positive rates. In contrast to the algebraic model of conics, we use the geometric model of ellipse and minimize the geometric distance of the fitted ellipse from the digital curve. As a result, the solutions are strictly restricted to ellipses. Results demonstrate a superior performance than most least squares based method for elliptic curves and greater true negative rates for non-elliptic curves even in presence of 30% noise.
Year
DOI
Venue
2012
10.1007/978-3-642-31295-3_30
ICIAR
Keywords
Field
DocType
geometric model,squares method,core process,non-elliptic conic,geometric distance,non-elliptic curve,ellipse detection,precise ellipse fitting method,noisy data,fitted ellipse,algebraic model,squares model
Least squares,Image processing,Artificial intelligence,Non-linear least squares,Ellipse,Mathematical optimization,Pattern recognition,Geometric modeling,Algorithm,Conic section,Elliptic curve,Mathematics,Constrained optimization
Conference
Volume
ISSN
Citations 
7324
0302-9743
2
PageRank 
References 
Authors
0.38
14
3
Name
Order
Citations
PageRank
Dilip K. Prasad116221.84
Chai Quek2143195.84
Maylor K. H. Leung3778135.09