On estimating the uncertainty in the location of image points in SD recognition from match sets of different sizes

Efficient and robust model-based recognition systems need to be able to est imate reliably and quickly the possible locations of other model features i n the image when a match of several model points to image points is given. Errors in the sensed data lead to uncertainty in the computed pose of the o bject, which in turn lead to uncertainty in these positions. We present an efficient and accurate method for estimating these uncertainty regions. Our basic method deals with an initial match of three points. With a small add itional computational cost it can be used to compute the uncertainty region s of the projections of many model points using the same match triplet. The basic method is then extended employing statistical methods to estimate un certainty regions when given initial matches of any size. This is the major practical contribution of the paper because when the number of points in t he match increases, the size of the uncertainty region decreases dramatical ly, which helps to discriminate much better between correct and incorrect m atches in model-based recognition algorithms. (C) 1999 Academic Press.