UAV photography in forest fire sites

We propose a GCP-free photography workflow for post-wildfire surveys in mountainous regions, where the implementation of ground control points (GCPs) is not feasible for rapid surveys. The workflow integrates three sets of imagery : close-range in-forest images acquired using an AI-assisted UAV (Skydio 2＋), forest-edge and open-area images obtained with a UAV equipped with network RTK GNSS (Mavic 3E), and bridging images collected to ensure overlap between the in-forest and open-area scenes. These three datasets are processed within a single Structure-from-Motion (SfM) and Multi-View Stereo (MVS) pipeline to propagate georeferencing to the in-forest imagery. Georeferenced imaging of post-wildfire conditions facilitates subsequent AI-based analyses and provides a contemporaneous record of site conditions. We estimate that the proposed workflow has the potential to reduce on-site fieldwork time from 2 hours and 31 minutes to 56 minutes, achieving a 62.9% reduction.