A Match Method of Encircled Marker-Points on External Store Model

Abstract

In the high-speed and complex flow circumstance, the recognition, location and matching of the encircled markerpoints on the aircraft external store model are the important basic data for determining the separation characteristics of the model and the flight trajectory. In order to solve the matching problem of the marker-points on the small-sized external store, a algorithm of encoding and decoding with encircling markerpoints is proposed. The decoding algorithm calculates the distance of marker-points between head and tail and the direction vector to match the homonymous points. Such algorithm has the advantages of keeping the aerodynamic shape, easy to implement, and low cost. The experiments show that the decoding accuracy is higher than 94% in circumstance of long distance and large field of view, complex background and high ambiguity. In addition, the efficiency of decoding is also improved. By applying the decoding with encircling marker-points in the monocular vision measurement system, the experiments show that the proposed method can achieve fast recognition and high-precision matching. It has engineering practical value.