MissionLabAirborneDataset-Clouds

MLAD-C The MissionLabAirborneDataset-Clouds contains aerial images of clouds and associated labeled cloud masks from a flight experiment conducted on October 12, 2023, in the Upper Bavaria region of Germany. The augmented dataset consists of 5488 RGB images captured from forward-looking perspective. MLAD-C is designed to develop models for cloud segmentation. In our cloud detection approach, cloud segmentation functions as a pre-stage to cloud position estimation for sense & avoid purposes. More detailed information about MLAD-C and the developed cloud segmentation model can be found in our journal article: A Cloud Detection System for UAV Sense and Avoid: Analysis of a Monocular Approach in Simulation and Flight Tests Directory Structure MLAD-C is provided in two formats: augmented sensor recordings prepared for YOLO-v8 framework to reproduce results of journal article non-augmented sensor recordings in Full HD resolution The directory structure is as follows: augmented_mladc train images labels masks val images labels masks mladc.yaml full_hd_mladc images masks Cite If you use MLAD-C, please cite the following publication: Dudek, A.; Stütz, P. A Cloud Detection System for UAV Sense and Avoid: Analysis of a Monocular Approach in Simulation and Flight Tests. Drones 2025, 9, 55. https://doi.org/10.3390/drones9010055 Funding The research project MissionLab is funded by dtec.bw – Digitalization and Technology Research Center of the Bundeswehr which we gratefully acknowledge. dtec.bw is funded by the European Union – NextGenerationEU.