A new model-based method of recognizing a moving space polygon from two images

Two novel methods are proposed. The first is used to find the altitude (or even the position if the polygon size is available) of a polygon in space from a single-perspective view. The second is used to recognize a moving polygon with two images. The first method is based on a numerical solution and has the following advantages compared to the other method: the computation in the search procedure for correct attitude is greatly reduced; the polygon shape is not limited to only a few regular figures; and prior knowledge of vortex correspondences between model and image is not necessary.<
>