抽象的

Performance of RANSAC Techniques under Classical and Robust Methods

R. Muthukrishnan, E.D. Boobalan, R.Reka

Robust statistics deals with deviations from the assumptions on the model and is concerned with construction of statistical procedure which still reliable and reasonably efficient in a neighbourhood of the model. In computer vision, a robust estimator should be able to correctly find the fit when outliers/noise occupies a high percentage of the data. RANSAC is most commonly used robust estimator in computer vision task. It is a nondeterministic algorithm for estimation of a mathematical model from observed data which contains outliers. This paper presents the performance of RANSAC algorithmic techniques under the various methods LS, LMS, LTS and M estimators along with the results of the simulation study which is carried out in MATLAB.