Research on UAV Visual Positioning Technology Based on Multi-Source Image Hierarchical Matching Mechanism

Visual geolocation technology is a hot issue in the field of UAV autonomous navigation. In the process of UAV flight, the change of geographical environment characteristics leads to the matching difficulty caused by the significant difference between aerial image and reference image, and the real-time performance cannot meet the application requirements. In this paper, a UAV visual positioning method based on multi-source image hierarchical matching mechanism is proposed. This method uses hierarchical matching to realize the mapping from aerial images to satellite images, and according to the position of the mapped images in the wide-area remote sensing reference map, realizes the transformation from the matching results of aerial images and remote sensing images to the geographical positioning results, and solves the problem of UAV visual positioning in the area where the geographical features of the images change.