Published February 23, 2021
| Version v1.0
Dataset
Open
Video stabilisation results obtained using different tools for UAS-based image velocimetry
Authors/Creators
- 1. Department of Hydraulic and Environmental Engineering, Faculty of Civil Engineering, University of Belgrade
- 2. Carinthia University of Applied Sciences, School of Geoinformation
- 3. School of Geography, Politics and Sociology, Newcastle University
- 4. Institute of Photogrammetry and Remote Sensing, Technische Universität Dresden
- 5. Photrack AG
- 6. Department of Civil, Architectural and Environmental Engineering, University of Naples Federico II
- 7. Department of European and Mediterranean Cultures: Architecture, Environment and Cultural Heritage (DICEM), University of Basilicata
Description
This dataset contains unstabilised and stabilised images from unmanned aerial systems during image velocimetry campaigns. Stabilised images were obtained using seven digital image stabilisation tools:
- FFT-CUAS (https://bitbucket.org/SIENA_Research/fishstream)
- FFT-DCH (https://github.com/salpeha/FFTVidStabilization)
- SSIMS (https://github.com/ljubicicrobert/SSIMS)
- KLT-IV (https://sourceforge.net/projects/klt-iv)
- VISION/FAST (https://doi.org/10.17605/OSF.IO/HBRF2)
- FlowVeloTool/AKAZE (https://github.com/AnetteEltner/FlowVeloTool)
- Blender (https://www.blender.org)
The tools have been tested with three case studies with different types and magnitudes of motion commonly found in UAS/UAV video footages:
- Kolubara river,
- Alpine river, and
- Basento river.
The structure of the repository is: /Case_study/Stabilisation_tool_results. To reduce the repository size, only images are included. In order to create videos from the provided images, one can use the provided Python script (for Python v3+).