A simple method for pose estimation of a circle, with known circle radius, on the basis of binocular stereo vision is presented. The method is based on the projective equation of a circle, into which the stereo vision constraints are incorporated. The estimated pose parameters are compared with the ground truth and excellent performance is reported.rnCircular features have long been used in model based computer vision because of their easy availability. Thus, it is an essential task to estimate the pose parameters of a circle for many machine vision applications, such as autonomous navigation and noncontact industrial detection. Safaee-Rad et al. [1] proposed a method for this task using binocular stereo vision. They applied a two-stage strategy, whose procedure is first to estimate the circle pose parameters, with redundant solutions, in the left and right camera co-ordinate system, respectively, and then to eliminate ambiguity by introducing constraints from stereo vision.
展开▼
