Effective line detection with error propagation

In this paper, we introduce a new Hough Transforni aimed at improving curve detection accuracy and robustness as well as coniputational eficiency. Robustness and accuracy improvement is achieved by analytically propagating the errors with image pixels to the estimated curve parameters. The errors with the curve parameters are then used to determine t k contribution of pixels to the accuniulator array. The coniputationai rjjiciency is achieved by choosing best-distinguished pixels and by performing progressive detection. The detection approaches were given for line and circle. The concept can be applied to other curves, such as circle and ellipse. The experiments on line detection show iniproved performance with our teclinique.