A RANSAC-based approach to model fitting and its application to finding cylinders in range data

摘要: General principles for fitting models to data containing "gross" errors in addition "measurement" are presented A technique is described and illustrated by its application the problem of locating cylinders range data, two key steps this process arc ellipses partial lines sets three-dimensional points The specifically designed filter out gross before applying a smoothing procedure compute precise model Such particularly applicable computer vision tasks because these often produced local computations that inherently unreliable.