A novel target shape recognition method for Aerial Robot based on minimum circle-cover matching

Target shape recognition is an important aspect of Aerial Robot (AR) information perception. A novel target shape recognition method for AR, which is based on minimum circle-cover matching, is proposed in this paper. The minimum circle-cover and diameter of the target shape are taken as shape feature, then minimum circle-cover and several scale factors are calculated to form the shape description of target profile, which is used to match with template profile. In this way, the target can be detected, and the designated target can be matched quickly and accurately. Simulation results demonstrated the effectiveness and feasibility of the proposed approach. This approach provides a practical way to solve the target shape recognition problems for AR effectively.