Error propagation for the Hough transform

In this paper. a statistically efficient Hough transform (HT) technique wit h improved performance in accuracy and robustness is described. The propose d technique analytically computes the uncertainty of each feature point bas ed on image noise, the procedure used for estimating edge orientation, and the specific parametric representation scheme of a line. Using the estimate d uncertainty of each feature point, a Bayesian probabilistic scheme is int roduced to compute the contribution of each feature point to the accumulato r. A performance evaluation of our technique reveals its improved performan ce, especially for noisy images. (C) 2001 Elsevier Science B.V. All rights reserved.