Real-time detecting and tracking method for autonomous landing landmark of unmanned aerial vehicle on the basis of vision
暂无分享,去创建一个
The invention discloses a real-time detecting and tracking method for the autonomous landing landmark of an unmanned aerial vehicle on the basis of vision. A detected target image is subjected to gridspot spreading, a pixel is used for replacing the target image, a pyramid L-K optical flow method is used for tracking, a situation that a feature point extraction algorithm is used for extracting feature points which can be easily tracked is avoided, so that time is saved, and algorithm instantaneity is improved. The bidirectional pyramid L-K optical flow method is adopted to track the feature points, and meanwhile, normalized correlation matching verification is adopted for each feature point to guarantee feature point tracking accuracy. A target tracked in a previous frame is taken as a template and is subjected to normalized correlation matching detection with the target image tracked in the current frame so as to guarantee the integrity and the accuracy of the tracking target of thecurrent frame.
[1] Emanuele Frontoni,et al. A Vision-Based Guidance System for UAV Navigation and Safe Landing using Natural Landmarks , 2010, J. Intell. Robotic Syst..