| Presentation | 2012/1/12 Accuracy Comparison of Ellipse Fitting : From Least Squares to Hyper-Renormalization KENTA YOKOTA, KAZUHIRO MURATA, YASUYUKI SUGAYA, KENICHI KANATANI, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We summarize the following techniques for fitting an ellipse to a point sequence extracted from an image: "least squares" and its update by "iterative reweight", the "Taubin method" and its iterative update by "renormalization", "HyperLS" and its iterative update by "hyper-renormalization", "maximum likelihood (ML)" which minimize the reprojection error and its a posteriori "hyperaccurate correction". We experimentally compare their accuracy and show the following: 1. Newly proposed hyper-renormalization is more accurate than ML, which has been widely regarded as the most accurate. 2. The most accurate is the hyperaccurate correction of ML, but the difference from hyper-renormalization is very small. 3. While iterations for computing ML may not always converge in the presence of large noise, Hyper-renormalization is more robust that ML. From these, we conclude that hyper-renormalization is the best method in practical situations. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | |
| Paper # | Vol.2012-CVIM-180 No.24 |
| Date of Issue |
| Conference Information | |
| Committee | CQ |
|---|---|
| Conference Date | 2012/1/12(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Communication Quality (CQ) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Accuracy Comparison of Ellipse Fitting : From Least Squares to Hyper-Renormalization |
| Sub Title (in English) | |
| Keyword(1) | |
| 1st Author's Name | KENTA YOKOTA |
| 1st Author's Affiliation | Department of Computer Science, Okayama University() |
| 2nd Author's Name | KAZUHIRO MURATA |
| 2nd Author's Affiliation | Department of Information and Computer Sciences, Toyohashi University of Technology |
| 3rd Author's Name | YASUYUKI SUGAYA |
| 3rd Author's Affiliation | Department of Information and Computer Sciences, Toyohashi University of Technology |
| 4th Author's Name | KENICHI KANATANI |
| 4th Author's Affiliation | Department of Computer Science, Okayama University |
| Date | 2012/1/12 |
| Paper # | Vol.2012-CVIM-180 No.24 |
| Volume (vol) | vol.111 |
| Number (no) | 378 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |