In this paper, we consider a new unmanned aerial vehicle (UAV)-assisted oblique image acquisition system where a UAV is dispatched to take images of multiple ground targets (GTs). To study the three-dimensional (3D) UAV trajectory design for image acquisition, we first propose a novel UAV-assisted oblique photography model, which characterizes the image resolution with respect to the UAV’s 3D image-taking location. Then, we formulate a 3D UAV trajectory optimization problem to minimize the UAV’s traveling distance subject to the image resolution constraints. The formulated problem is shown to be equivalent to a modified 3D traveling salesman problem with neighbourhoods, which is NP-hard in general. To tackle this difficult problem, we propose an iterative algorithm to obtain a high-quality suboptimal solution efficiently, by alternately optimizing the UAV’s 3D image-taking waypoints and its visiting order for the GTs. Numerical results show that the proposed algorithm significantly reduces the UAV’s traveling distance as compared to various benchmark schemes, while meeting the image resolution requirement.
UAV-Assisted Image Acquisition: 3D UAV Trajectory Design and Camera Control
01.09.2022
1126816 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Acquisition of Pedestrian Trajectory around Vehicle Using On-board Camera
British Library Online Contents | 2011
|TRAJECTORY DESIGN FOR IMAGE DATA ACQUISITION FOR OBJECT DETECTION/RECOGNITION
Europäisches Patentamt | 2023
|Trajectory design for image data acquisition for object detection/recognition
Europäisches Patentamt | 2024
|