Fitting ellipses to noisy measurements

TL;DR

This paper reviews various methods for fitting ellipses to noisy data, compares their performance through simulations, and discusses robust techniques for handling outliers, providing open-source code for reproducibility.

Contribution

It provides a comprehensive comparison of ellipse fitting methods, including robust approaches for outlier resistance, with simulation results and publicly available code.

Findings

Robust methods outperform traditional ones in the presence of outliers

Simulation results highlight the strengths and weaknesses of different approaches

Open-source code facilitates further research and application

Abstract

This work deals with fitting of ellipses to noisy measurements. The literature knows many different approaches for this. The main representatives are presented and discussed in this paper. Furthermore, the case is considered when outliers are present in the measurement data. Robust methods which are less sensitive to outliers are suitable for this case. All discussed methods are compared by a simulation. The code for the simulation is available for free use on github.com/sebdi/ellipse-fitting.