Abstract
This paper proposes a fast algorithm for computing the real roots of univariate polynomials given in the Bernstein basis. Traditionally, the polynomial is subdivided until a root can be isolated. In contrast, herein we aim to find a root only to subdivide the polynomial at the root. This subdivision based algorithm exploits the property that the Bézier curves interpolate the end-points of their control polygons. Upon subdivision at the root, both resulting curves contain the root at one of their end-points, and hence contain a vanishing coefficient that is factored out. The algorithm then recurses on the new sub-curves, now of lower degree, yielding a computational efficiency. In addition, the proposed algorithm has the ability to efficiently count the multiplicities of the roots. Comparison of running times against the state-of-the-art on thousands of polynomials shows an improvement of about an order-of-magnitude.
| Original language | English |
|---|---|
| Pages (from-to) | 16-26 |
| Number of pages | 11 |
| Journal | Computer Aided Geometric Design |
| Volume | 43 |
| DOIs | |
| State | Published - Mar 2016 |
Keywords
- Bezier polynomials
- Newton-Raphson, polynomial division
- Numerical method
- Polynomial roots
- Subdivision
ASJC Scopus subject areas
- Modelling and Simulation
- Automotive Engineering
- Aerospace Engineering
- Computer Graphics and Computer-Aided Design
Fingerprint
Dive into the research topics of 'Revisiting the problem of zeros of univariate scalar Beziers'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver