Georeferenced image stitching with pose optimization using image similarity with little overlap

Abstract

The generation of high-quality ortho-images using unmanned aerial vehicles is essential for precision agriculture. For quick surveys with little overlap between images, georeferenced aerial image mosaicking methods are effective because they have shorter processing times than 3D reconstruction-based methods. However, it is difficult to apply them in areas where the correspondence of feature points cannot be obtained using existing stitching methods. A high-quality image stitching method with pose optimization using image similarity and feature point correspondence has been proposed. This method has been experimentally demonstrated to be a feasible and promising solution for the generation of high-quality ortho-images for quick surveys.